Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats





Volume XXXIII July, 1939 



Application of Motion Picture Film to Television 


A Continuous Type Television Film Scanner. . P. C. GOLDMARK 18 

Television Studio Technic A. W. PROTZMAN 26 

Television Lighting W. C. EDDY 41 

An Introduction to Television Production. . . .H. R. LUBCKE 54 

Design Problems in Television Systems and Receivers 

A. B. DuMoNT 66 

Report of the Television Committee 75 

Properties of Lamps and Optical Systems for Sound Reproduc- 
tion F. E. CARLSON 80 

Report of the Studio Lighting Committee 97 

Report of the Projection Practice Committee 101 

Report of the Committee on Exchange Practice 103 

Report of the Membership Committee 106 

Sound Picture Recording and Reproducing Characteristics .... 

Current Literature , 109 

1939 Fall Convention, New York 112 





Board of Editors 
J. I. CRABTREE, Chairman 




Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1939, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


** President: E. A. WILLIFORD, 30 East 42nd St., New York, N. Y. 
** Past-President: S. K. WOLF, RKO Building, New York, N. Y. 
** Executive Vice-President: N. LEVTNSON, Burbank, Calif. 

* Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
** Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 

* Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 
** Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 90 Gold St., New York, N. Y. 

* Treasurer: L. W. DAVEE, 153 Westervelt Ave., Tenafly, N. J. 

** M. C. BATSEL, Front and Market Sts., Camden, N. J. 

* R. E. FARNHAM, Nela Park, Cleveland, Ohio. 

* H. GRIFFIN, 90 Gold St., New York, N. Y. 

* D. E. HYNDMAN, 350 Madison Ave., New York, N. Y. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* S. A. LUKES, 6427 Sheridan Rd., Chicago, 111. 

** H. G. TASKER, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1939. 
** Term expires December 31, 1940. 


Summary. Motion picture film will form an important source of programs for 
television broadcasting. Film projectors for this use are required to meet a number of 
conditions peculiar to television. Methods for projecting and utilizing motion picture 
film are outlined. A specific film projector and associated television channel are de- 
scribed in some detail. 

In establishing a technicfor producing films most suitable for television, equipment 
is needed to interpret the final results. Apparatus that will be used by broadcasting 
stations is described. A simpler system has been designed that may be useful for the 
specialized service of gauging the merit of films for television. This is described and 
its operation indicated. 

Some very preliminary observations are included on the characteristics of films that 
have given good results in experimental work and infield tests. 

The production and utilization of motion picture film for television 
programs introduce many new problems. It is the purpose of this 
paper to review these problems and to describe methods and appara- 
tus for the use of film in television. 

It is desirable first to review the general characteristics of two elec- 
tronic television pick-up systems that are known to give practical 
results. In both systems the scene to be transmitted is projected 
upon a photoemissive area or mosaic. The resulting "electrical 
image" is methodically explored by electronic means, one narrow 
strip or line at a time, in a process called scanning. The result of this 
scanning process is an electrical signal which varies in accordance 
with the scene brightness along the scanning lines. The information 
residing in this signal is used at the receiver to reconstruct the image 
one element at a time in a similar synchronized scanning process. 

In one pick-up system, exemplified by equipment using the Farns- 
worth dissector tube, only the light falling upon an element of the 
photoemissive area at the instant that element is being scanned is 
effective in producing the signal. The other pick-up system, ex- 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received May 8, 

** RCA Manufacturing Co., Catnden, N. J. 



[J. S. M. P. E. 

amplified by equipment using the Iconoscope, makes use of the princi- 
ple of storage, whereby, when a particular photoemissive element is 
scanned, the light that has fallen upon that element since it was last 
scanned is effective in producing the signal. 

The characteristics of these pick-up tubes determine the manner in 
which film can be used to provide television programs. In the system 
using the dissector tube, which has no storage, for every instant the 
signal is transmitted the film projector must supply a light image to 
the elemental area being scanned, though not necessarily from the 

FIG. 1. Schematic of film projector for Iconoscope camera. 

entire frame. In the Iconoscope system, utilizing storage, a charge 
image may be built up by a very brief projection of the image upon 
the photoemissive mosaic, which is then scanned by an electron beam 
while the mosaic is dark, to produce the signal. The film pull-down 
occurs during the relatively long interval while the mosaic is being 
scanned. The detailed discussion to follow will be based on the sys- 
tem utilizing the Iconoscope. 


Fig. 1 shows schematically an Iconoscope camera and a special 
projector adapted to project standard 24-frame-per-second film upon 
the Iconoscope mosaic in such way as to generate television signals 


according to the Radio Manufacturers Association standards: 
namely, at 30 frames per second and 60 fields per second, interlaced. 1 
The projector must flash a still picture upon the mosaic every Veo 
second with each flash lasting less than Veoo second. Since the film 
must run at a mean speed of 24 frames per second for proper repro- 
duction of sound and motion, it is evident that each frame must be 
projected more than once to provide the required sixty flashes per 
second. Since sixty divided by 24 is 2 1 /*, it would seem logical that 
each frame should be projected two and one-half times. This is 
impracticable but a very satisfactory method is to project alternate 
frames of film two and three times each, respectively; for example, 
the even frames twice and the odd frames three times. Fig. 2 shows 
the various steps of projection and scanning in proper relative time 
on a horizontal time-scale. Since the light flashes are very brief, a 

i.M FRAME N &%%% FILM FRAME N+ t KtZZA N + 2 


FIG. 2. Preferred sequence of events in film transmission by Iconoscope. 

relatively long (approximately 1 / N second) interval is available be- 
tween flashes for the film pull-down. However, if the full time avail- 
able is used, the alternate pull-downs must occur at non-uniform 
intervals of 2 /eo and 3 /eo second, respectively. Note from this figure 
that the scanning or transmission times occur between adjacent light 
flashes so that the television picture signal is actually produced and 
transmitted during periods when no optical image is present on the 
mosaic. However, during these periods an electrical image is present 
in the form of bound electrostatic charges on the tiny photosensitized 
silver globules comprising the mosaic. It is the act of neutralizing or, 
rather, equalizing these charges by the electrons of the scanning 
beam that causes the useful signal current to flow from the conducting 
back-coating of the mosaic plate. 

Referring again to Fig. 1, the film is drawn through an illuminated 
gate by an intermittent sprocket which is driven by an intermittent 
cam and spider-follower of the early Powers type. The 3600-rpm 
special synchronous motor drives the cam at 12 revolutions per second 


through a suitable gear, thus pulling the film down 24 times per 
second, since the cam has two "throws" instead of the customary one 
"throw." In order to pull the film at unequal intervals as required, 
the "throws" are located 144 degrees and 216 degrees apart, respec- 
tively. The film picture in the gate is projected upon the small 
photoemissive mosaic of the Iconoscope by a standard projection 
lens. The light is chopped 60 times per second by a large rotating 
shutter, located near the lens. The shutter is accurately timed rela- 
tive to the intermittent cam so that the film is always stationary when 
the light flashes occur. 

The generator of synchronizing signals for the television deflecting 
system is synchronously controlled by the same 60-cycle power supply 
that drives the projector synchronous motor. The phase of this 
signal generator is adjustable so that the operator can make the 
short-duration light flashes fall safely within the Veoo-second intervals 


FIG. 3. Idealized sequence of events in film transmission by Iconoscope. 

between the vertical scanning periods with some tolerance on each 
side for slight phase displacements such as are caused by small 
changes in the mechanical load on the projector or by voltage varia- 
tions. This adjustment is very important, as any abrupt change in 
the illumination of the mosaic during the picture signal transmission 
time produces a spurious light streak across the received picture. 
An ordinary 3600-rpm synchronous motor has two identical pole 
structures which can assume either polarity, and hence such a motor 
can lock into synchronism in either of two phase positions, depending 
fortuitously upon starting conditions. Two such lock-in positions 
are apart in time by one-half of a cycle of the power-supply frequency, 
which for a 60-cycle power system is l /m second. Inspection of the 
diagram of Fig. 2 shows that displacing the light flashes Vi2o second 
with respect to the scanning periods would cause them to occur during 
instead of between the scanning periods. The abrupt change in 
mosaic lighting caused by a flash during the scanning period would 
produce a serious streak across the middle of the picture as mentioned 

July, 1939] 


above. To prevent the frequent locking-in of the motor in the wrong 
position, a special synchronous motor is used which includes an addi- 
tional d-c winding for fixing the polarity of the poles and thus de- 
termining the lock-in position with respect to the a-c power supply. 

The sound-head used is standard, since the mean speed of the film is 
24 frames per second. It has been found that a suitable fly-wheel 

71/ TIME - 7* 

FIG. 4. Experimental rocking mirror projector. 

associated with the intermittent cam prevents any detectable deteri- 
oration of the reproduced sound due to the dissymmetry of the inter- 
mittent cam. 


There is some evidence that the television picture transmitted by a 
system depending completely upon the storage principle might not be 
as satisfactory as one transmitted by a system in which the film image 



is projected upon the photoemissive mosaic either continuously or 
during the entire scanning period. It is natural, therefore, that in- 
vestigations of the latter type of system should have been made. 
So far, the results obtained have not been wholly satisfactory and 
certainly have not been as excellent as those produced by the storage 
method described in the previous section. However, refinement of 
certain projection methods may at some time in the future make other 
systems of greater interest. It is, therefore, of value to digress and 

review some of the various 
schemes that have been inves- 

For obtaining a continuous 
and constant light image on 
the Iconoscope photoemissive 
mosaic, a commercial type of 
theater projector was used, hi 
which the film passed the pic- 
ture gate at constant speed 
and a stationary projected 
image was obtained by means 
of an "optical intermittent." 
This projector employed 
several rocking mirrors on a 
rotating wheel. The lens sys- 
tem was properly proportioned 
for the projection of the small 
image required for the Icono- 
scope mosaic plate. In testing 
this system it was noted that 
the television performance was 
limited by various types of movement in the projected optical image 
and by low resolution. Motion of the optical image, hi addition to 
causing objectionable motion in the received television picture, also 
contributed to loss of resolution in the picture. This is due to the 
storage action of the Iconoscope whereby the signal derived from 
each element of the mosaic in scanning is due to the summation of all 
the light that has fallen on that element since the preceding contact 
of the scanning beam. The effect is similar to that obtained when 
the optical image on a sensitized photographic plate moves during 

FIG. 5. RCA 35-mm sound motion 
picture projector; 30 frames per second, 
interlaced scanning. 


Fig. 3 shows a projection sequence by which an intermittent type 
projector might project film on an Iconoscope for the entire scanning 
time provided the pull-down occurred in the almost prohibitively 
short time of l /m second or less. This would permit projection 
throughout the entire scanning period. There is no apparatus now 
available for meeting the Veoo second pull-down requirement. If 
suitable equipment could be developed it is doubtful whether the 
film would withstand the stresses imposed by the rapid motion. 

FIG. 6. Film projector, with doors open. 

An experimental projector using a continuously moving film, and a 
rocking mirror for producing a stationary image, was built and tested. 
A diagrammatic view of it is shown in Fig. 4. The cam-driven mirror 
was arranged to neutralize accurately the film motion during the 
intervals marked "light flash" in Fig. 3 and to return to receive light 
from the next consecutive film frame during the Yew-second non _ 
uniformly spaced intervals marked "pull-down." Limitations were 
found due to slight non-uniform illumination of the approximately 
two and one-half frames of film always in the picture gate. This 
resulted in objectionable flicker in the television picture. Also, in 
spite of the very small amplitude of motion required for the rocking 



[J. S. M. P. E. 

mirror, the cam and follower-roller created a very annoying noise and 
were subject to rapid wear. 


It is of interest to return now to the method for using film which is 
considered best at present, and review the apparatus in more detail. 

FIG. 7. View of film projector, showing film path. 

Fig. 5 is a general view of a 35-mm sound motion picture projector 
designed for 30-frame-per-second television with interlaced scanning. 
This projector differs from standard theater projectors in the following 
major respects : 

(1) A special shutter is used to provide efficient light pulses of very short time 
duration for projecting, 60 tunes per second, images of the film pictures onto the 
photoemissive mosaic of the Iconoscope. 



(2) The intermittent mechanism is designed for the three-to-two ratio of 
pull-down periods required in using 24-frame film for 30-frame television. 

(5) A special synchronous driving motor is used to assure that the projector 
mechanism always "locks-in" in proper time relation with the synchronizing 

(4) An additional film gate with light-source and photoelectric cell is included 
near the picture gate for deriving a control potential which varies with the average 
density of the film. 

In the projector shown in Fig. 5, it was impracticable to locate the 
shutter between the light-source and the film. The shutter was, 
therefore, mounted just beyond the projection lens. Sufficient 
clearance between the shutter and lens was provided to permit limited 
movement of the lens for focusing. The time during which the image 






FIG. 8. 

Essential elements of a television film transmis- 
sion system. 

may be projected, onto the photoemissive mosaic of the Iconoscope 
is limited to the vertical return time of the scanning beam. With 
present television standards this is not more than 10% of Veo second 
or Veoo second. 

In order to make efficient use of the projection lens, it is necessary 
that the aperture in the shutter be at least as wide as the diameter of 
the lens. A large-diameter shutter (23 inches) is necessary to meet 
this requirement. This shutter rotates at 3600 rpm and has a peri- 
pheral speed of approximately 4y 4 miles per minute. The shutter is 
enclosed in the circular housing shown at the extreme right-hand side 
of Fig. 5. In the shutter housing opposite the projection lens is a 
window through which the picture is projected. The shutter disk is 
made of two overlapping sections of thin metal. These two sections 
can be rotated with respect to each other through a small angle in 


order to vary the width of the aperture. Fig. 6 is a photograph show- 
ing the film side of the projector with the cover removed. 

A second gate is located four frames of film above the picture gate. 
To the left of this gate, as shown in Fig. 6, is a lamp housing. To the 
right of this gate is a photocell housing which also includes an optical 
system for forming an image of the lamp filament on the photocell. 
The output voltage from this photocell is rectified, and after being 
passed through a suitable filter is used to control a characteristic of 
the synchronizing signals. The resultant variation in the synchron- 
izing signals is used to control the average brightness of the reproduced 
picture. Fig. 7 shows a view of the film side of the projector with a 
film threaded ready for projection. Although the projector just 
described is equipped with a small 30-ampere arc, either an incandes- 
cent lamp or an arc may be used. 


In considering the production of motion picture films for television, 
it is important to review the apparatus that will be used in the 
broadcasting station. The essential elements of a system for tele- 
vision transmission from motion picture film are shown in Fig. 8. 
These include : 

Film Projector Control Equipment 

Iconoscope Film Camera Monitor Equipment 

Camera Amplifier Equipment Synchronizing Generator 

The Iconoscope camera used with the film projector includes de- 
flecting circuits and a pre-amplifier for the video signals. This pre- 
amplifier provides a signal level suitable for transmission over a 
coaxial cable to the camera amplifier equipment. The camera is 
usually mounted on one side of a wall, with the film projector located 
on the other side. The picture is projected through a window in the 
wall into the camera onto the photoemissive mosaic of the Icono- 

The camera amplifier equipment includes apparatus for amplifying 
further the video signals from the camera and a line amplifier to 
prepare these signals for transmission over coaxial cable to any de- 
sired location. Amplifiers providing suitable wave-shapes for hori- 
zontal and vertical deflection of the Iconoscope beam are included 
as well as the power supplies for the several parts of the system. 
This equipment is usually rack-mounted in some convenient location. 


The control equipment provides means for varying the video signal 
gain, the picture brightness, and the picture background illumina- 
tion, and for starting and stopping the film projector. In an installa- 
tion designed to provide a continuous program from motion picture 
film, where two or more film projectors and television channels are 
included, controls are also provided for switching from one channel to 
another. The monitor equipment includes a 12-inch Kinescope by 
means of which television images obtained from the film can be 
viewed. It includes also a cathode-ray oscilloscope for observing 

FIG. 9. Television control equipment for studio and film type 

the wave-shapes and amplitudes of the television signals. This 
monitor equipment is usually located so that it may be observed 
conveniently by the operator manipulating the control apparatus. 

The synchronizing generator supplies the several complex wave- 
forms required to determine the timing of scanning processes in the 
transmitting equipment and to synchronize the reconstruction of the 
images at the receivers. The wave-shapes of the synchronizing 
signals have been standardized by the Radio Manufacturers Associa- 

Views of television equipment of a type suitable for television 
broadcasting stations are shown in Figs. 9 and 10. Fig. 9 shows an 
installation of control equipment for studio and film type cameras. 


This equipment is grouped on a common control console with the 
monitors mounted in a recess in the wall above the console. In this 
installation, the control engineer may look directly into the studio. 
Fig. 10 shows a typical installation of racks of television terminal 


For specialized services, more simple and compact television equip- 
ment is desirable. Apparatus of this sort has been developed both 

FIG. 10. Terminal equipment for television broadcasting stations. 

for direct studio pick-up and for film applications. The simplified 
equipment suitable for producing television signals and television 
images from motion picture film will be reviewed briefly. 

This apparatus includes all the elements previously described, but 
in far more compact form. The equipment less the Iconoscope 
camera and the projector is included in one cabinet approximately 44 
inches high, 34 inches wide, and 21 inches deep. This equipment pro- 
duces a television signal that is suitable for transmission to remote 
viewing positions or for other uses. 

This simplified equipment is not as flexible in some respects as the 
broadcasting type of equipment, nor does it lend itself well to large, 



complex systems. However, it does provide the facilities necessary 
for judging the merits of film for television use. In this simplification 
of apparatus and circuits, the synchronizing wave-shapes do not con- 
form entirely to the Radio Manufacturers Association standards. 
The synchronizing signals are, however, satisfactory for the self- 
contained monitor and for other receivers or reproducing devices, 
but the adjustments may be a little more critical than would be the 
case with standard synchronizing signals. Fig. 1 1 shows a view of the 
equipment with the Iconoscope camera mounted on a simple wooden 

FIG. 11. Simplified television apparatus for 
judging the merits of motion picture film. 



An earlier paper 1 reviewed some of the limitations inherent in pres- 
ent-day television and compared them with similar limitations in 
motion picture film and apparatus. Experience has indicated that 
the production of television pictures from a particular film is the only 
practicable method for judging the merits of the film as television 
program material. It is therefore suggested that this method be used 
for checking and studying motion picture films produced for television 
programs and for determining the usefulness of film available from 


other sources. Apparatus of the type used at the television broad- 
casting station or apparatus of the simplified type just described will 
be satisfactory for this service. 


Laboratory work and field test experience permit some prelimi- 
nary generalizations on film that has given good results for television. 
Comment is here directed to the technical characteristics of film and 
not to the entertainment qualities. It appears that film having 
characteristics best suited for theater projection is also generally 
best for television. Studio sets having all dark backgrounds should 
be avoided. A good number of close-ups should be used but these 
should be generously interspersed with long shots. Some experience 
may be necessary to take into account the resolution limits 1 of pres- 
ent-day television. Special processing of film does not seem to be 

Film photographed in color directly from real life or nature appears 
satisfactory for television. Some cartoons in color have not given 
particularly satisfactory results. 

Thus, it appears that there may be no really serious technical 
problems in the production of motion picture films suitable for tele- 
vision program material. 


1 BEERS, G. L., ENGSTROM, E. W., AND MALOFF, I. G.: "Some Television 
Problems from the Motion Picture Standpoint," /. Soc. Mot. Pict. Eng., XXXII 
(Feb., 1939) p. 121. 


MR. MACKEOWN: Why has television adopted 30 frames a second when 24 are 
used in ordinary moving pictures? 

MR. ENGSTROM : To explain fully the choice of 30 frames for television would be 
beyond the scope of the present discussion. Papers were published in the JOURNAL 
of this Society in January and February of this year that dealt with this particular 

Basically the reason is that electron beams are used for scanning at the trans- 
mitter to produce the television signals, and electron beams for scanning at the 
receiving or reproducing end. These electronic devices are affected by fluctua- 
tions in supply voltages and stray variations in electric and magnetic fields. The 
visual result is a pattern of varying brightness with respect to time that corre- 
sponds to the extraneous influence. If the frame frequency is made a whole- 
number submultiple of the power-supply frequency, then the visual patterns in 
the reproduced image are stationary and the effects are much less pronounced. 


If this "whole-number submultiple" relationship does not obtain, then patterns 
of light and shade will travel up or down the image at a rate corresponding with 
the difference rate between the frame frequency and the nearest submultiple of 
the power supply. These moving patterns are as objectionable as true flicker. 

If we did not use 30-frame television, for 60-cycle power, the designer would be 
confronted with very severe handicaps. For receivers, in particular, it would be 
difficult to find operative apparatus layouts, and certainly the cost would be in- 
creased to provide tolerable freedom from the effects just outlined. 

MR. CRABTREE: At the New York meeting you demonstrated a method of pro- 
jecting on a fairly large screen an image of high brightness contrast. Would it 
not be better to photograph the televised image onto motion picture film, process 
that rapidly, and then project it? What are the relative merits of the two pro- 

MR. ENGSTROM : I believe the method you suggest has been experimented with 
in Germany and to the best of my knowledge has not been adopted. One of the 
advantages that television has is the timeliness of the reproduction and it appears 
to those who have thought about the subject that even a matter of twenty, thirty, 
or forty seconds would take much interest from the program. On the other hand, 
to photograph the image on motion picture film and then project it involves a 
great deal of apparatus. We hope to find a more direct and simpler answer which 
will make it more generally applicable. 

MR. CRABTREE: I do not think that a time lag detracts seriously in all cases 
from the showmanship. We in the East are quite entertained in spite of an ap- 
parent three-hour delay in the transmission of the football game from California 
on New Year's Day. 

MR. GOLDEN: Is there any difference in definition between receiving film 
images and images of living actors? 

MR. ENGSTROM : If as good a job is done in the television studio as one would 
do in producing motion picture film, then the resulting performance should be 
identical. There is no reason why one system should permit better contrast or 
better resolution than the other. 

MR. LUBCKE: Our results concerning the qualifications for good film for tele- 
vision support those that Mr. Engstrom has outlined. In addition, we have felt 
that a slightly accentuated contrast between the various parts of the scene, but 
not too great an overall contrast range, is also desirable. 


Summary. A motion picture film scanner, the first of the continuous type to be 
used for television transmissions, is described. The apparatus was put into operation 
in New York City in the summer of 1937 and has been in use since. In its preferred 
form the scanner projects the image of a continuously moving film onto the cathode of a 
dissector tube. Five images, representing different portions of the film in the gate, 
produced by five stationary lenses, are superimposed one on top of the other, while a 
rotating shutter with concentric slots permits only one lens at a time to produce an im- 
age. The scanning is accomplished partly by the uniform motion of the film and 
partly by the magnetic scanning of the electron image in the opposite direction. The 
pictures thus obtained are completely free from shading, cover a great range of contrast, 
are free from flicker, and are steady. The construction of the scanner is simple and 

While the Iconoscope is the best device known at present to pick up 
scenes where the amount of illumination is limited, the same tube is 
not the best for the transmission of motion picture films. Devices 
like the scanning disk, dissector tube, projection type cathode-ray 
tube, etc., will produce pictures of greater contrast and freedom from 
the shading effects that represent one of the greatest disadvantages of 
the Iconoscope. Film scanners which project a continuous visual 
image onto the Iconoscope or onto a dissector tube have been sug- 
gested and are in use. However, they all involve rotating parts such 
as lenses or mirrors, these parts in some cases numbering as many as 
forty-eight, which must be preadjusted with extreme accuracy, mak- 
ing subsequent corrections at the projector nearly impossible. Jt was 
therefore desirable to develop equipment in which the optical ele- 
ments involved would be stationary and few in number. 

Another requisite for such a film scanner is that the film should 
move through the gate with constant speed instead of with jerky 
motion, as it does hi the intermittent type of projector, so that the 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 7, 

** Columbia Broadcasting System, New York, N. Y. 




film perforations can be maintained in good condition over a longer 
period of time. 

The film transmitter under discussion employs no moving optical 
elements and operates on the following principle : The motion picture 
film has to pass a gate at a speed of 24 frames per second while one 
frame has to be scanned electrically within Veo of a second. This 
means that within Via of a second two complete motion picture 
frames have to be scanned electrically five times. 

In Fig. 1, frame A, which has just emerged from the top of the gate, 
is moving downward at a constant rate of 24 frames per second. At 
the same time a fictional scanning-light spot starts to scan the same 

FIG. 1. Illustrating passage of film in gate. 

frame from the bottom upward. After Veo of a second the scanning 
spot describing horizontal lines will have covered a height of 3 /6 f 
a frame, which, plus a downward 2 /5 motion of the frame, equals the 
picture height. Position B indicates how far frame A has moved in 
the gate within Veo of a second while it was scanned once from 
bottom to top. The scanning-beam returns now to the base-line of 
that frame, and while the frame continues from position B toward 
position C, the scanning spot, again in Veo of a second, covers that 
frame. Now the frame continues to move from C to D, which trip 
again lasts Veo of a second, and the scanning spot quickly returns to 
the base-line of C and moves upward, completing the third frame. 
By that time a new frame, E, appears in the gate. The scanning- 



[J. S. M. P. E. 

beam instead of further describing frame D, continues to scan frame E 
which is on its way down from position E to F. Again the scanning 
spot returns to the base-line of frame F with a rapid jump, while this 
frame proceeds to position G and is scanned from bottom to top within 
Vso of a second. Thus an entire cycle has been completed, consist- 
ing of the scanning of two motion picture frames five times each 
within Veo of a second, or the total operation completed in Vi2 of 
a second. 

The scanning spot will now start to scan frame H which appears at 

the top of the gate, and thus start a 
new cycle. 

This principle can be applied in 
several ways in practice, some of 
which will be mentioned here. A 
luminous scanning pattern can be pro- 
duced on the flat end of a high- 
intensity cathode-ray tube which is 
projected through lenses onto the 
moving film. With a suitable con- 
denser lens system, light emerging 
through the film is collected and pro- 
jected into a multiplier type of photo- 
cell. An extra pah- of deflecting coils 
or deflecting plates fed by a rec- 
tangular current or voltage displaces 
the luminous scanning pattern on the 
tube in such a fashion that various 
portions of the moving film are scanned 
hi accordance with Fig. 1. Although 
this method is theoretically correct, 
hi practice difficulties are encountered due to the fact that in order to 
obtain five congruent images, the scanning pattern as it jumps into 
various positions on the screen of the tube must maintain its shape 
within Viooo of its height and width. It is very difficult to produce a 
sufficiently homogeneous magnetic and electrical field that will pro- 
duce identical images. 

The thought occurs to produce a stationary scanning pattern and 
split it up into five images displaced in accordance with Fig. 1. To 
split up the original into five displaced images we can employ 
either five mirrors tilted at different angles or five lenses with their 

FIG. 2. Schematic layout of 
scanner using five mirrors. 

July, 1939] 



optical axes displaced, or parallel blocks of glass inclined toward the 
main optical axis at suitable angles. 

The field-frequency of 60 frames per second conforms to the U. S. 
television standards and requires that five images be produced and 
displaced within Vi2 of a second. A rotating shutter with five con- 
centric slots, each slot covering an angle of 72 degrees and a disk 
rotating at a synchronous speed of 720 rpm permits the passage of 
light for only one image at a time. 

Fig. 2 is the schematic layout of a scanner employing five mirrors 
for splitting up the beam, while Fig. 3 shows a similar layout 


FIG. 3. Schematic layout using five lenses: (a) ar- 
rangement of optical parts; (b) virtual images and dis- 
placement of real images by the -five dissecting lenses. 

using five lenses. The total amount of light available at the multi- 
plier photocell when splitting up the original image is l /& of that 
obtainable with the first method where the optical or electron image 
itself is deflected and no image-splitting takes place. However, there 
is sufficient light available for both methods when a suitable in- 
candescent lamp is used. 

Fig. 4 is a photograph of the latest film scanner using a dissector 
tube. In this arrangement a 1500-watt projection lamp is used as 
light-source which illuminates the gate evenly. In Fig. 3 the mirror 
M and the condenser lens system L produce an intermediate image of 
the filament in the plane x-x. This image is projected through the 

22 P. C. GOLDMARK [J. S. M. P. E. 

condenser system c\ in combination with field lens F and the main 
projection lens T, onto the plate of the five dissecting lenses, LI to L 6 . 
The main projection lens T produces an enlarged virtual image of the 
film behind the gate in the plane y-y. The five lens segments displace 
and project this enlarged image of the gate onto the cathode of the 
dissector tube, the displacement of the images corresponding to 
the displacement of the centers of the dissecting lenses relatively to the 
main optical axis of the entire system. Fig. 3(b) reveals the virtual 
images and the displacement of the real images by the five dissecting 

FIG. 4. Film scanner using dissector tube. 

The five lens sections are rigidly mounted on a solid metal plate and 
are easily adjustable with the aid of small brackets and screws holding 
the lenses in place. The alignment of the dissecting lenses can be 
carried out quickly by the following method. 

A reel of film of a suitable resolution chart is run through the pro- 
jector while two lens sections are freed by using cardboard masks over 
the others. The two images thus produced on the cathode are 
brought into coincidence by adjusting the lens-holder and observing 
the monitored image on a cathode-ray tube. The other three lenses 
are brought into coincidence with the first pair one after the other, 

July, 1939] 



and the brackets holding the lenses are permanently clamped down. 
With this method coincidence of the five images can be obtained 
accurate to within a fraction of the width of a line both in the hori- 
zontal and the vertical directions. It is, of course, necessary that the 
film running continuously through the projector be steady within 
these limits. However, suitable mechanical filters make such uni- 
form film speed possible. 

There is one more important consideration in this film-scanning 
method that requires close attention. The distances between centers 
of the five dissecting lenses in the vertical direction correspond to 
certain distances between successive film-image centers on the film. 

FIG. 5. Diagram of complete telecine channel employing the projector. 

At any distance between the plane of the dissecting lenses and the 
cathode of the pick-up tube sharp focus of the film picture together 
with most accurate coincidence of the five images must be available. 
However, it is known that there are rarely two reels of film that 
exhibit the same shrinkage, that is, the same spacing between centers 
of successive images. A method had to be developed whereby the 
system could be quickly adjusted for any shrinkage while maintaining 
good focus and coincidence over the entire image. It has been deter- 
mined experimentally that the shrinkage of film will vary anywhere 
between and 12 mm per 50 frames, which corresponds roughly to 
from to 0.8 per cent. A large number of test-reels with suitable 
patterns, each of which showed different shrinkage between the limits 

24 P. C. GOLDMARK [J. S. M. P. E. 

mentioned before, were run through the projector. Each time 
the distance between the dissector cathode and a fixed point on the 
projector was carefully adjusted until the images were motion- 
less and in sharp focus. Through suitable gears the slight changes 
of the dissector distance are transmitted to a rotating dial onto which 
the individual positions of the cathode have been marked. From this 
calibration a curve has been derived which represents the film shrink- 

FIG. 6. Electrical equipment of the telecine channel. 

age (measured in mm per 50 frames) plotted against the distance be- 
tween the dissector cathode and the projector. 

In practice, before a film of unknown shrinkage is used, a piece of it 
is held against a metal ruler on which the shrinkage can be read 
directly in mm per 50 frames. Then the distance adjustment screw S 
(Fig. 4) of the dissector is turned until the dial D of the indicator 
reads the corresponding distance. Ah 1 that is necessary now is to 
focus the main projection lens T of the projector, when the title ap- 
pears and automatically both a sharp image and exact coincidence are 


Refocusing during a picture is seldom necessary. The depth of 
focus of the main projector lens combined with the five lens segments 
(the magnification introduced by the latter is 1:1) is large compared 
with the tolerances in the movement of the main projector lens so far 
as coincidence of the individual images is concerned, so that a slight 
focusing movement of the main lens, by means of a knob K (Fig. 4) 
brought out conveniently on the side of the projector, will always 
bring the image to exact coincidence and substantially not change the 
optical focus. 

A diagram of the entire telecine channel employing this projector 
is shown in Fig. 5. Fig. 6 shows the electrical equipment, com- 
prising, from right to left, synchronizing-pulse generator rack, 
power-supply rack, video amplifiers and scanning generator rack, 
and, finally, the picture and wave-shape monitor as well as the 
blanking-pulse amplifier rack. This telecine system is operated on 
the so-called d-c principle. The picture background component is 
accurately maintained in the cathode of the pick-up tube by means of 
horizontal and vertical blanking pulses which are injected between the 
cathode and the first multiplier stage of the dissector. 

Thus the peaks of the blanking pulses, representing black, maintain 
a constant reference level all through the system. Wherever a visual 
picture is needed, the a-c axis of the video signal is so changed by one 
of several well known methods that the tops of the blanking pulses 
are all along a straight line establishing a definite black level in any 
kind of picture. 

Two years of experience with the film-transmitting system just 
described have brought results that compare favorably to those ob- 
tained by the scanning disk method employed abroad. It is believed 
that this system is one of the simplest of all film-scanning devices 
permitting continuous and reliable operation without readjustments 
or breakdown. 

The writer wishes to acknowledge the helpful assistance of his asso- 
ciates and especially of Mr. Bernard Erde of the CBS Television 


Summary. The studio operating technic as practiced in the NBC television 
studios today is discussed and comparisons are made, where possible, to motion 
picture technic. Preliminary investigations conducted to derive a television operating 
technic revealed that both the theater and the motion picture could contribute certain 

The problems of lighting, scenic design, background projection, and make-up are 
discussed, with special emphasis on the difficulties and differences that make television 
studio practice unique. 

An explanation is given of the functioning of a special circuit used in television 
sound pick-up to aids in the creation of the illusion of close-up and long-shot sound 
perspective without impracticable amount of microphone movement. The paper con- 
cludes with a typical television production routine showing the coordination and timing 
of personnel and equipment required in producing a television program. 

If one were forced to name the first requirement of television oper- 
ating technic and found himself limited to a single word, that word 
would undoubtedly be "timing." Accurate timing of devices and 
split-second movements of cameras are the essentials of television 
operation. Personnel must function with rigid coordination. Mis- 
takes are costly they must not happen there are no second 

Why such speed and coordination? Television catches action at 
the instant of its occurrence. Television does not allow us to shoot 
one scene today and another tomorrow, to view rushes or resort to 
the cutting room for editing. Everything must be done as a unit, 
correct and exact at the time of the "takes" otherwise, there is no 
television show. 

Now, to discuss some preliminary investigations conducted before 
production was attempted, and to describe the equipment and tech- 
nic used hi meeting these production requirements. Technical de- 
tails are deliberately omitted. Wherever possible, we shall compare 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 6, 

** National Broadcasting Co., New York, N. Y. 


phases of television operation with their counterparts in motion pic- 
ture production. 

For so new a medium as television it is, of course, an impossibility 
to present a complete and permanently valid exposition. Television 
technic and apparatus constantly advance. Some technic now cur- 
rent may be outmoded in a day or a month. We have only to recall 
the early days of motion picture production, when slow-speed film and 
inferior lenses were a constant limitation. So, with television, it is 
already possible to envision more sensitive pick-up tubes that will 
permit the use of smaller lenses of much shorter focal length, thus 
eliminating many of today's operating difficulties. 

Production Technic Investigations. In May, 1935, the Radio Cor- 
poration of America released television from its research laboratories 
for actual field and studio tests. Long before the first program was 
produced in the middle of 1936, plans were laid, based on extensive 
research into the established entertainment fields, for the purpose of 
determining in advance what technics might be adaptable to the new 
medium of television. From the stage came the formula of con- 
tinuity of action, an inherent basic requirement of television. This 
meant memorized lines and long rehearsals. Prompting could not be 
considered, for, as you know, the sensitive microphone which is as 
much present in television as it is in sound motion picture production, 
does not discriminate between dialog and prompting. 

From the motion picture studio came many ideas and technics. 
If television is a combination of pictures with sound, and it is, no 
matter what viewpoint is taken, the result spells in part and for many 
types of programs, a motion picture technic at the production end. 
However, enough has already been said about the peculiarities of 
television presentation to justify saying that the movie technics do 
not supply the final answer. There remained the major problem of 
preserving program continuity without losing too much of motion 
picture production's flexibility. Our present technic allows no time 
for adjustments or retakes. Any mistake immediately becomes the 
property of the audience. The result of the entire investigation led 
to what we think is at least a partial answer to the problem. This 
technic, we hope, will assist considerably in bringing television out 
of the experimental laboratory and into the field of home education 
and entertainment. 

General Layout of Facilities. In order to present a clearer view of 
our problems, we shall give a brief description of our operating plant. 



[J. S. M. P. E. 

The present television installation at the National Broadcasting Com- 
pany's headquarters in the RCA Building, New York, N. Y., consists 
of three studios, a technical laboratory, machine and carpenter shops, 
and a scenic paint shop. Our transmitter is located on the 85th floor 
of the Empire State Building. The antenna system for both sight 
and sound is about 1300 feet above the street level. Both the 
picture and sound signals are relayed from the Radio City Studios to 
the video and sound transmitters either by coaxial cable or over a 
special radio link transmitter. 

FIG. l(a). General layout of live- talent studio; control room at upper rear. 

One of the studios is devoted exclusively to televising motion pic- 
ture film, another to programs involving live talent, and the third for 
special effects. It is the operation of the live-talent studio with which 
we are concerned in this paper. 

Description of Live-Talent Studio. Fig. 1 (a) shows the general layout 
of the live- talent studio. The studio is 30 feet wide, 50 feet long, and 
18 feet high. Such a size should not be considered a recommendation 
as to the desired size and proportions of a television studio. The 
studio was formerly a regular radio broadcasting studio, not espe- 

July, 1939] 



daily designed for television. To anyone familiar with the large 
sound stages on the motion picture lots, this size may seem small 
(Fig. !(&)). Yet, in spite of our limited space, some involved multi- 
set pick-ups have been successfully achieved by careful planning. 
Sets, or scenes, are usually placed at one end of the studio. Control 
facilities are located at the opposite end in an elevated booth, afford- 
ing full view of the studio for the control room staff. Any small sets 
supplementing the main set are placed along the side walls as near 
the main set as possible, and in such position as to minimize camera 
movement. At all times, we reserve as much of the floor space as 


FIG. 1(6). Television studio floor plan. 

possible for camera operations and such floor lights as are absolutely 
essential. At the base of the walls and also on the ceiling are scat- 
tered numerous light-power outlets to minimize the length of lighting 
cables. At the rear of the studio is a permanent projection room for 
background projection. 

Camera Equipment. The studio is at present fitted for three 
cameras. To each camera is connected a cable. This cable is about 
two inches in diameter and fifty feet long; it contains 32 conductors 
including the well known coaxial cable over which the video signal is 
transmitted to the camera's associated equipment in the control room. 
The remainder of the conductors carry the necessary scanning volt- 

30 A. W. PROTZMAN [j. s. M. P. E. 

ages and current supplies for the camera amplifiers, interphone sys- 
tem, signal lights, etc. From this description, it is apparent that 
adding another camera in a television studio involves a much greater 
problem than that of moving an extra camera into a motion picture 
studio. In television, it is necessary to add an extra rack of equip- 
ment in the control room for each additional camera. 

Movement of Cameras. One camera, usually the long-shot camera 
using a short focal length lens, is mounted on a regular motion picture 
type dolly to insure stable movements. The handling of the dolly is 
done by a technician assisting the camera operator. It is impracti 

FIG. 2. Studio camera. 

cable to lay tracks for dolly shots as is often the motion picture prac- 
tice, because usually each camera must be moved frequently in all 
directions during the televising of a studio show. Naturally, dolly 
tracks would limit such movement. The other television cameras 
utilize a specially designed mobile pedestal (Fig. 2). Cameras 
mounted on these pedestals are very flexible and may be moved in 
and out of position by the camera operators themselves. Built into 
the pedestals are motors which elevate or lower the camera; this ac- 
tion is controlled with push-buttons by the camera operators. A 
panning head, similar to those used for motion picture cameras, is also 


a part of the pedestal. It is perhaps needless to stress here that one 
of the strict requirements of a television camera is that it must be 
silent in operation. In the electronic camera proper there are no 
moving parts other than those used for focusing adjustments; hence, 
it is a negligible source of noise. When camera pedestals were first 
used they were the source of both mechanical noise and electrical dis- 
turbance when the camera-elevating motor was in use. Since then 
this problem has been overcome, and it can be stated that the entire 
camera unit is now free of objectionable mechanical noise or electrical 

Lens Complement: Each camera is equipped with an assembly of 
two identical lenses displaced 6 inches vertically. The upper lens 
focuses the image of the scene on a ground-glass which is viewed by 
the camera operator. The lower lens focuses the image on the "mo- 
saic," the Iconoscope's light-sensitive plate. This plate has for its 
movie counterpart the film in a motion picture camera. The lens 
housings are demountable and interchangeable. Lenses with focal 
lengths from Q l /z to 18 inches are used at present. Lenses of shorter 
focal length or wider angle of pick-up can not be used since the dis- 
tance between the mosaic and the glass envelope of the Iconoscope is 
approximately 6 inches. Lens changes can not be effected as fast as 
on a motion picture camera, since a turret arrangement for the lenses 
is mechanically impracticable at present. However, it is probably 
safe to say that future advances in camera and Iconoscope design will 
incorporate some type of lens turret. Ordinarily, one camera utilizes 
a 6V2-inch focal length lens with a 36-degree angle, for long shots, 
while the others use lenses of longer focal lengths for close-up shots. 
Due to its large aperture, the optical system used at present has con- 
siderably less depth of focus than those used in motion pictures, mak- 
ing it essential for camera operators to follow focus continuously and 
with the greatest care. This limitation will probably be of short 
duration, since more sensitive Iconoscopes will permit the use of opti- 
cal systems of far greater depths of focus. 

It is desirable here to point out a difference in focusing technic be- 
tween motion picture cameras and television cameras. "Follow- 
focus" in motion pictures occurs practically only in making dolly 
shots. For all fixed shots, the lens focus is set, the depth of focus 
being sufficient to carry the action. Also, it is the duty of the assistant 
cameraman to do the focusing. This relieves the cameraman of that 
responsibility and allows him to concentrate on composition, action, 

32 A. W. PROTZMAN [j. s. M. P. E. 

and lighting. In television, the camera operator must do the focusing 
for fixed shots and dolly shots alike. This added operation, at times, 
is quite fatiguing. 

Vertical parallax between the view finder lens and the Iconoscope 
lens is compensated for by a specially designed framing device at the 
ground-glass that works automatically in conjunction with the lens- 
focusing control. It may be of interest to note here that at first the 
television camera had no framing device. This meant that images, in 
addition to being inverted as they are in an ordinary view-finder, were 
also out of frame. The camera operator had to use his judgment in 
correcting the parallax. With this new framing device, the operator 
now knows exactly the composition of the picture being focused on 
the mosaic in his camera. The framing device can be quickly ad- 
justed to accommodate any lens between 6*/2 and 18 inches focal 

Because of the fact that several cameras are often trained on the 
same scene from various angles, and because all cameras are silent in 
operation, performers must be informed sometimes such as when 
they are speaking directly to the television audience which camera 
is active at the moment. Two large green bull's-eye signal-lamps 
mounted below the lens assembly are lighted when the particular 
camera is switched "on the air." 

Set Lighting. There are two outstanding differences between tele- 
vision lighting and motion picture lighting. A much greater amount 
of key light is required in television than in motion pictures. Also, a 
television set must be lighted in such a way that all the camera angles 
are anticipated and properly lighted at one time. Floor light is held 
to a minimum to conserve space in assuring maximum flexibility and 
speed of camera movements. Great care must also be taken to 
shield stray light from all camera lenses. This task is not always 
easy, since, during a half-hour performance, each camera may make 
as many as twenty different shots. Just as excessive leak-light 
striking the lens will ruin motion picture film, it has a definitely in- 
jurious effect upon the photosensitive mosaic and upon the electrical 
characteristics of the Iconoscope. A direct beam of high-intensity 
light may temporarily paralyze a tube, thus rendering it useless for 
the moment. 

Sets. (Fig. 3) Television sets are usually painted in shades of gray. 
Since television reproduction is in black and white, color in sets is 
relatively unimportant. Chalky whites are generally avoided be- 

July, 1939] 



cause it is not always possible to keep "hot lights" from these highly 
reflective surfaces which cause a "bloom" in the picture. This, in 
turn, limits the contrast range of the system. Due to the fact that 
the resolution of the all-electronic system is quite high, television sets 
must be rendered in considerable detail, much more, in fact, than for 
a corresponding stage production. As in motion picture production, 
general construction must be as real and genuine as possible; a 
marked difference, for instance, can be detected between a painted 

FIG. 3. Typical television set. 

door and a real door. On the legitimate stage, a canvas door may be 
painted with fixed highlights; that is, a fixed perspective, because the 
lighting remains practically constant, and the viewing angle is ap- 
proximately the same from any point in the audience. But, in tele- 
vision the perspective changes from one camera shot to another. 
Painted perspectives would therefore be out of harmony with a realis- 
tic appearance. This is also true in motion picture work. Sets must 
also be designed so that they can be struck quickly with a minimum 
effort and noise because it is often necessary to change scenes in one 

34 A. W. PROTZMAN [J. S. M. P. E. 

part of a studio while the show is going on in another part. At pres- 
ent, we find it desirable to construct television sets in portable and 
lightweight sections without sacrificing sturdiness. 

Background Projection. The problems of background projection 
in television differ somewhat from those encountered in motion pic- 
tures. More light is necessary because of the proportionately greater 
incident light used on the sets proper (Fig. 4). 

Considering the center of a, rear-screen projection as zero angle, we 

FIG. 4. Background projection window shot. 

must make it possible to make television shots within angles of at 
least 20 degrees on either side of zero without appreciable loss of pic- 
ture brightness. This requirement calls for the use of a special 
screen having a broader viewing angle than those used in making mo- 
tion picture process shots. Also, in motion pictures, the size of the 
picture on the screen can be varied to the proper relation to the fore- 
ground for long shots or close-ups. For television, the background 
picture size can not be changed once the program starts. Our back- 


ground subject matter must also be sharp in detail and high in con- 
trast for good results. 

At present, only glass slides are used. A self -circulating water-cell 
is used to absorb some of the radiant heat from the high-intensity arc. 
Also both sides of the slide are air-cooled. These precautions permit 
the use of slides for approximately 30-minute periods without damage. 

Make- Up. This may be a suitable time to correct some erroneous 
impressions concerning the type of make-up used in television. It 

FIG. 5. The television control room. Note the two 
Kinescope monitors in the upper left corner. 

has never been necessary to use gruesome make-up for the modern all- 
electronic-RCA television system. At present, No. 26 panchromatic 
base, similar to that used for panchromatic film, and dark red lipstick 
is being used satisfactorily. From the very beginning, we have made 
tests to determine the proper color and shades of make-up, keeping in 
mind that a color closely approximating the pigmentation of the hu- 
man skin is most desirable from the actor's psychological standpoint. 
The Control Room. Now, a few words about the operations in the 

36 A. W. PROTZMAN [j. s. M. P. E. 

studio control room during a televised production (Fig. 5). All 
camera operators in the studio wear head-phones through which they 
receive instructions from the control room. Directions are relayed 
over this circuit by the video engineer or the production director. 
Here the televised images are observed on special Kinescope monitors 
and necessary electrical adjustments are made. Alongside each of 
these monitoring Kinescopes is a cathode-ray oscilloscope which 
shows the electrical equivalent of the actual picture. Two monitors 
are provided in order that one may be reserved for the picture that is 
actually on the air, while the other shows the succeeding shot as picked 
up by a second or third camera. This enables the video engineer to 
make any necessary electrical adjustments before a picture goes on 
the air. 

Seated immediately to the left of the video engineer is the produc- 
tion director whose responsibility corresponds to that of the director 
of a motion picture. He selects the shots and gives necessary cues 
to the video engineer for switching any of the cameras into the out- 
going channel. The production director has, of course, previously 
rehearsed the performance and set camera routines in conjunction 
with the camera operators and the engineering staff. The camera 
operator has no control to switch his camera on the air. All camera 
switches, which are instantaneous, are made by electrical relays con- 
trolled by buttons in the control room. At present, the video engi- 
neer's counterpart in motion picture work is the editor and the film 
processing laboratory. 

To the left of the production director sits the audio control engineer 
whose responsibility is entirely separate from that of the video engi- 
neer. He also is in a position to view the monitor, and may com- 
municate by telephone with the engineer on the microphone boom. 
The audio engineer is responsible for sound effects, some of which are 
dubbed in from records. His job is somewhat similar to that of the 
head sound engineer on a motion picture production. Thus, we have 
the control room staff three men who have final responsibility for 
the success of the completed show. 

An assistant production man is also required on the studio floor. 
Wearing headphones on a long extension cord, he is able to move to 
any part of the studio while still maintaining contact with the pro- 
duction director in the control room during a performance on the air. 
Actors require starting cues, titles require proper timing, and proper- 
ties and even an occasional piece of scenery must be moved. The 


assistant director supervises these operations and sees that the in- 
structions of the production director are properly carried out. 

Members of the studio personnel also to be mentioned include 
lighting technicians, the property man, and scene shifters, whose re- 
sponsibilities parallel those of their motion picture counterparts. 
Specially trained men are also needed for operating title machines. 
In the future all titling will undoubtedly be done in a separate studio 
inasmuch as operating space in a television studio is at a premium. 
Today, however, title machines do operate in the studio and require 
the utmost care in handling. Types of titles used include dissolves 
and wipes similar to those used in moving pictures. 

Sound Reproduction. As in motion picture work, a microphone 
boom is used in television production, and is operated in a similar 
way. Perspective in motion picture sound is accomplished by keep- 
ing the microphone, during a long shot, just out of the picture and 
moving it down closer to the action as the camera moves in for a 
close-up, thus simulating a natural change in perspective. In tele- 
vision this is not always possible because there are always three 
cameras to consider. This same condition prevailed in the early days 
of motion pictures when it was thought desirable to take a complete 
scene, shooting both long-shot and close-up cameras, at one time. 
In the television studio at least one camera is always set for a long 
shot while the others are in position for closer shots. If the micro- 
phone is placed in such a position as to afford a "natural" perspective 
for close-ups, the succeeding switch to a long shot would reveal the 
microphone in the shot. You in motion pictures can order a retake ; 
in television broadcasting we can not rectify the mistake. It is quite 
obvious, therefore, that the man on the boom can not lower his micro- 
phone to the "natural" position for each camera shot. We therefore 
place the microphone in a position just out of range of the long shot. 
In order to accomplish some sense of perspective between long and 
close-up shots, a variable equalizer that drops the high and low ends 
of the spectrum is automatically cut into the audio circuits when the 
long-shot camera is on the air. In this operation, sufficient change in 
quality and level is introduced to aid the illusion of long-shot sound 
perspective. Of course, when a close-up camera is switched in, the 
audio returns to the close-up perspective quality once more. This 
may be called remote control sound perspective. 

Special sound effects, music, etc., from the studio picked up from 
recordings are mixed in the control room. In motion pictures, some 



[J. S. M. P. E. 

of the effects and most of the music are dubbed in after the actual 
shooting of the scene. 

The general acoustical problems in a television studio are similar to 
those in a motion picture sound-stage. Walls and ceiling should be 
designed for maximum absorption to permit faithful exterior speech 
pick-up. A stage or studio must be designed to enable presentation 
of an exterior or an interior scene. With the studio designed for 
maximum absorption, illusions of exterior sound characteristics can 
be created. For interiors, the hard surfaces of the sets and props 
offer sufficiently reflective surfaces to create the indoor effect. 

FIG. 6. (Left} Scene on the air; (right} setting up for next scene. 

Typical Production Routine. After the foregoing discussion of the 
equipment and personnel, it may be interesting to follow an actual 
production from the beginning of rehearsal to its final presentation. 
For this example, assume that we are to produce a playlet (Fig. 6). 
When the scenery has been erected, the first rehearsals begin without 
the use of cameras or lights. Besides familiarizing the actors with 
thetr lines, the rehearsals afford the production director and the head 
camera operator an opportunity to map out the action of the play. 


All action, including camera shots, cues, and timing, is noted on a 
master script which thereafter becomes the "bible" of the production. 
Timing is very important because of the necessity of having a particu- 
lar act time in with the other acts or film subject. 

After several hours of rehearsing, the first equipment rehearsal is 
called. Cameras are checked electrically and mechanically. Focus 
controls and framing devices are lined up so that correct focus on the 
ground-glass is also correct focus on the mosaic plate. This com- 
pleted, the cameras are ready for rehearsal. With the scene properly 
lighted, the camera operators begin working out movements to pick 
up the desired shots in the proper sequence. The production director 
instructs the staff and personnel from the control room, speaking over 
a public-address system. Each shot is worked out and its camera lo- 
cation marked on the floor. At times, the actors may unconsciously 
depart slightly from the rehearsed routine during an actual show; 
the camera operator must be prepared and alert to make the best of 
the situation regardless of all previous floor markings. Continuity 
is so planned that while one camera is taking the action, another 
camera is moving to a new location and composing a new shot to be 
switched on at the proper time. This frees the first camera, which 
can now move to a third location, and so on. Sometimes during a 
twenty-minute performance each camera may take twenty different 
shots. Of course, besides different floor locations, the height and 
angle of the cameras must be varied to comply with good composi- 
tion. During rehearsals, timing must frequently be revised to allow 
for the actual camera movements. 

Finally, a dress rehearsal is scheduled. The complete program is 
televised, including any film subjects or slides that may be needed to 
complete the program. Frequently the program will begin with a 
short film leader, followed immediately by a newsreel or a short sub- 
ject, the film portion of the program coming from the film-televising 
studio. While the film is running, the live- talent studio is continu- 
ously warned as to the time remaining before it must take over the 
program. Once the studio program goes on the air the production di- 
rector is no longer able to use the public address system to communi- 
cate with the personnel in the studio. Instead, he uses a telephone 
circuit to his assistant in the studio, and, through the video engineer, 
communicates by phone with the camera operators. 

Another standby warning is usually given when there is one minute 
to go. Then, as the cue to begin comes, the green light on the title 


camera is lighted. From this point, continuity must be rigidly pre- 
served. As titles move from one to another, appropriate music is 
cued in and actors are sent to their opening positions. 

With the completion of titles, the image is faded out electrically 
and cameras are switched to the opening shot. Performers begin 
their action on a silent cue from the assistant director, who is in- 
structed from the control room. During this first scene, the camera 
previously picking up titles moves quickly into position to shoot a 
second view of the action. Again cameras are switched, permitting 
the first to move to a new position ; and so the action proceeds. If 
the play has several scenes, the concluding shot of the first scene is 
taken by one camera while others line up on the new scene and wait 
for the switch. Frequently, there are outdoor scenes. These are 
filmed during the first stages of rehearsal for transmission from the 
film studio at the proper time during the performance. The switch 
to film is handled exactly as another camera switch, except that the 
switch is to the film studio instead of to one of the studio cameras. 
The projectionist must be warned in advance to have his projector up 
to speed and "on the air" at the proper instant to preserve the pro- 
duction continuity. This requires very critical timing, as you can 
well appreciate. When the film is completed the studio cameras 
again take over the next interior scene. 

Upon completion of the studio portion of the program, one camera 
lines up on the final studio title, which usually returns the program to 
the film studio for a concluding film subject. 

Since the first program on July 7, 1936, many television programs 
have been produced. Each has been a serious attempt at something 
new. Although much has been accomplished, there remain a vast 
number of unknowns to be answered before it can be said that tele- 
vision's potentialities have been even partially realized. Today, as 
this paper has indicated, television bears many points of similarity to 
motion pictures. As a matter of fact, it is likely that television 
would be somewhat handicapped if it were unable to borrow heavily 
from a motion picture production technic that has been built up by 
capable minds and at great expense over a period of many years. 
Infant television is indeed fortunate to have such a wealth of informa- 
tion at its disposal. Possibly continued experimentation will lead us 
toward a new technic distinctive of television. During its early years, 
however, television must borrow from all in creating for itself a book 
of rules. The first chapter of that book is scarcely written. 


Summary. Lighting a television production presents many problems peculiar to 
this new field of public entertainment. These problems have necessitated the redesign 
of lighting equipment and the establishment of a simplified technic for handling the 
equipment that differs radically from moving picture practice. 

To cope properly with the lighting requirements of the continuous action sequences, 
characterizing television productions, a system employing inside silvered incandescent 
lamps in a standardized unit was developed by NBC engineers. Based on multiple 
standardized group of l l /z kw each, these units are used in both the foundation light 
and modeling equipment of the television studios in Radio City, thus insuring quanti- 
tative as well as qualitative control of lighting by the personnel. 

With cameras generally in motion and an average duration of pick-up from one 
camera a matter of seconds, the problem of modeling in the sets becomes acute. This 
appears to be satisfactorily solved by the technic now in use wherein the major interest 
is centered around the close-up camera. Even this solution, however, required new 
and ingenious equipment to maintain light in the sets and still give floor precedence 
to the cameras and sound equipment. 

While NBC at the present time has appeared to have standardized on the inside 
silvered lamp, exhaustive tests were carried out in an attempt to utilize more orthodox 
equipment. Actual tests under production conditions proved, however, that certain 
requirements of space, weight, and flexibility could not be had without a serious sacrifice 
of foot-candles on the set, resulting in the present set-up of equipment and personnel 
that are handling the television lighting assignment in the East. 

Under these circumstances, our producers relying on their scientific skill, the 
richness of their facilities and resources, and the variety and range of talent available 
to them in every field will, it would seem, be well advised to stress most strongly in 
the foreign markets the factor of the superior quality of American films . We should 
export only pictures of unquestioned excellence. High quality will continue to retain 
for American motion pictures an exceedingly worth-while place in the markets of the 

Although the practical application of lighting to the presentation 
of television studio programs will admittedly be subject to further 
improvement, the imminence of a public television service warrants 
a description of the lighting equipment and operating technic which 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 13, 1939. 

** National Broadcasting Co., New York, N. Y. 


42 W. C. EDDY [J. S. M. P. E. 

the National Broadcasting Company has worked out as a result of 
several years of experimentation in this field. 

This description covers primarily the lighting developments since 
1935 when the Radio Corporation of America launched an extensive 
experimental field-test of television. Of considerably greater scope 
than previous tests, it was designed to permit a pre-commercial anal- 
ysis of the art through a combined appraisal of the laboratory- 
reared electrical system and a comprehensive survey of the problems 
introduced by regular production of programs. 

Starting with studio lighting equipment similar to that used in 
moving pictures, we have gradually evolved a reasonably satisfactory 
solution of our illumination problem that has resulted in a new and 
interesting layout of equipment applicable to the demands imposed 
by television studio operation. This was achieved largely through 
simplification of the equipment and the technic involved in handling 

To permit both engineer and director to discuss the lighting set-up 
with a common terminology, and thus facilitate presentations, we 
also simplified the existing abstract definitions of light into two sepa- 
rate and distinct classifications: namely, foundation and modeling 

Foundation light, according to our standards, is the non-character- 
istic flat illumination of a set, irrespective of its origin or amount. It 
is primarily the light energy necessary to create an electrical picture 
in the cameras and provide a foundation to which we can add the 
characteristic or dynamic quality of modeling light. 

Modeling light is any illumination that adds to the contrast or de- 
lineation of the picture. It may be from overhead, from the floor, 
or from the back, but according to our definition, it must create some 
characteristic highlight or shadow, as opposed to the flat illumination 
function of the foundation lights. 

It was, then, the creation of a satisfactory lighting installation for 
television rather than the adaptation of equipment and technics 
geared to an older art that paced our developmental work. It may 
help to follow the reasoning behind our transition from motion picture 
lighting into the present installation of incandescent sources, if we 
consider chronologically the television studio work at Radio City dur- 
ing the formative period from 1935 to the fall of 1938. 

A rough analysis of the requirements for a satisfactory system 
seemed to indicate that flexibility and efficiency were the paramount 

July, 1939] 



FIG. 1 (Upper}. A stage set-up in the television studio. 
FIG. 2 (Lower}. View of studio showing equipment. 

factors to be considered, although glare and radiant heat from the 
units had to be taken into account. Of necessity, the light produced 
had to be a high-level diffused illumination in quantities encountered 
only in the color-film studios. In addition, television required that 
the operation, upkeep, and maneuvering of this light be of such sim- 
plicity that one or two men could satisfactorily handle routine pro- 

44 W. C. EDDY [J. S. M. P. E. 

ductions. We naturally turned to the standardized fixtures of the 
moving picture lots for our first tests. In the Radio City studio we 
installed routine spots and broads. Due to the limitation of a nine- 
teen-foot ceiling, a practical light bridge was out of the question. As 
a substitute, the major portion of our lighting equipment was in- 
stalled on portable stands. Figs. 1 and 2 show the arrangement of 
the apparatus for our first television program from Radio City in 
1936. From a quantitative standpoint, we had little to criticize in 
this installation, but it was immediately apparent that the excessive 
glare and operational requirements of such a battery of lights pre- 
cluded their general use in television. An attempt was made to re- 

FIG. 3. Illumination by battery of 500-watt units. 

design and redistribute these units, but with little or no success, in- 
dicating conclusively that equipment of such power and concentra- 
tion could not be left unattended throughout a television sequence and 
that the proper manipulation of this type of illumination required a 
lighting personnel of considerable magnitude. 

Our next step was a gradual conversion from the concentrated type 
of unit to the more diffused and uniform light produced by scoop re- 
flectors and floor broads. Focusing spots and suns were still main- 
tained in the studio, but their function was limited to modeling rather 
than producing the foundation illumination. Lack of space for opera- 
tion, weight, and their general inefficiency coupled with unbearable 
glare on the set soon proved their impracticability even though the 


unattended light produced by high-efficiency lamps met the require- 
ments of the production staff. During this period, little attempt was 
made to do more than spill into the sets a predetermined quantity of 
shadowless light lacking the characteristic modeling that might prove 
embarrassing in certain sequences. Such a technic reduced the person- 
nel to a minimum, to be sure, but it also produced a television pic- 
ture in the field that was flat, non-dimensional, and on the whole, 
highly unsatisfactory from the program standpoint. 

Our next experimental step toward a television lighting system came 
with the installation of a battery of 500-watt units (Fig. 3), each 
equipped with separate reflector and lens systems. These lights 

FIG. 4. The single-six mounting. 

were positioned on a gridiron over a single set in such a manner that 
they would produce a cube of uniform, nondirectional illumination 
that, it was hoped, would approximate the character and modeling 
obtained under high-intensity diffused light. Needless to say, the 
resultant picture showed the effect of flat front lighting. Again the 
spots and suns were brought out the storeroom and put into opera- 
tion as modeling units in an attempt to create above this pedestal of 
1500 foot-candles the highlights and shades that had been destroyed 
by the basic arrangement of the foundation-light installation. Be- 
cause this system of multi-unit lighting was the first radical departure 
from orthodox lighting practice and the forerunner of our present 



[J. S. M. P. E. 


studio equipment, it might be well to go into more detail concerning 
its advantages and shortcomings. 

Coupled with the failure of this installation to produce the required 
quality of light were several equally important deficiencies : namely, 
lack of flexibility, excess weight, and great heat radiation. By reason 
of the bulk of the single unit alone it was necessary to select a certain 
area to be illuminated, a limitation that required the program group 
to parade their subjects within the confines of a limited stage. This 
placed a definite limitation on the efforts of this program group. The 
weight of the installation closely approached the safe load limit of our 

acoustical ceiling, making impos- 
sible the addition of further 
equipment above the set to rein- 
force the existing light or to create 
special light effects. The unit 
inefficiency of each lamp, lens, 
and exterior reflector created an 
ambient heat problem that se- 
verely taxed the air-conditioning 
service to this particular studio. 
These deficiencies made the adop- 
tion of this system inadvisable 
but did indicate the direction of 
our next step. 

Photometric tests, conducted 
in the studio, have already indi- 
cated that the new inside silvered 
spotlight would deliver into an 
area more light per watt than 

the lens, lamp, and reflector assembly or the standard incandescent 
bulb and exterior scoop. This new bulb was light in weight and of 
relatively small envelope size in the wattage required. It remained 
to design a fixture that would permit simple adjustment in elevation 
and direction to satisfy the requirements of the multi-set productions 
proposed by the program staff. Fig. 4 shows such a mounting, 
known as the "single six." It incorporates six 500-watt spotlights on 
a framework of thin-walled steel tubing, so arranged that the center- 
to-center distance between lights is ten inches. This insures that the 
light-beams interlock at a distance of eight feet from the fixture and 
that the light arriving on the set is relatively free from spots and 

FIG. 5. Light distribution curves of 
single-six unit and lens reflector unit. 

July, 1939] 



FIG. 6. Photometric distribution 
of the beam about the center-line. 

secondary shadows. The total 
weight of the fixture, equipped with 
spots, is slightly less than 19 pounds 
and lamped for three kilowatts pro- 
duces an index of 18,000 units, com- 
pared with an index of 7650 units 
registered by an equivalent grouping 
of lens, lamp, and reflector units. 
Roughly, this amounts to an in- 
crease in usable light per watt con- 
sumed of approximately 240 per 
cent. The distribution of these two 
test fixtures is best demonstrated by 
referring to polar coordinate curves 
projected on an area of approxi- 
mately 200 square-feet from a height 
of eight feet. In Fig. 5 the 300- 
foot-candle intensity curve for the 

"single six" is indicated by the solid line; that of the competitive fix- 
ture is shown dotted. Areas within these limits serve to indicate 
relative efficiencies, as the wattage, arrangement, and length of throw 
were held constant in obtaining the data. Fig. 6, with the solid line 

again indicating the "single six," 
gives a general idea of the photo- 
metric distribution of the beam 
about the center line. 

The mechanical arrangement 
for flexibility consists of a uni- 
versal clamp for attaching the 
supporting arm to a gridiron, with 
rotational freedom possible at the 
fixture itself. A single adjusting 
screw allows the operator to set 
the bank for any desired angle 
or direction of throw with the 
framework arranged either hori- 
zontally or in a vertical position 
relative to the studio floor. 

The first of the standardized in- 
FIG. 7. The double-three unit. stallations consisted of eighteen of 


W. C. EDDY- 

[J. S. M. P. E. 

these "single-six" units mounted on the gridiron in such a manner 
that they could quickly and easily be brought into play on any acting 
area selected by the production group. As a space-conserving meas- 
ure a few of these long units were reassembled in two rows of three 
(Fig. 7), designated as "double threes." In certain sets where the 
light-concentration was high and space at a minimum, this arrange- 
ment was found to be more satisfactory from an operational stand- 
point. This type of construction was later mounted on portable 
stands for use as floor broads. 

FIG. 8 (Left). The single-three unit. 
FIG. 9 (Right). The floor broad. 

The "single three" (Fig. 8), one-half of the "single six," was next 
brought into use for reinforcing light, background flooding, and as a 
general-purpose strip-light of minimum dimensions. 

By standardizing the construction of our unit assembly we were 
assured of uniform spectral characteristics and distribution from each 
fixture rather than a spotty heterogeneous mixture of several types 
of light requiring careful blending on the set. A common standard 
of light-producing unit also allowed us to familiarize ourselves with 

July, 1939] 



the operation of the fixture and, by simple addition or subtraction, to 
meet the studio's quantitative light problems. 

Shortly after completing the foundation-light installation we turned 
to the more complex problem of supplying the characteristic, or model- 

FIG. 10. Portable foot light. 

ing, light from the floor. Here again, several problems confronted us, 
resulting in a partial redesign of the standardized mounting 

The floor broad (Fig. 9) is identical with the overhead array except 
that it is mounted on a portable floor stand. Two of these units are 
used normally as reinforcing lights from stage right and left to create 
a rough modeling angle or to 
temper the shadows on the back- 
drops. In all cases, however, it 
was required that the operation 
of these lights should give floor 
precedence to camera movement. 
They are, therefore, brought 
into play and taken out fre- 
quently during the course of a 
single sequence. The diffused 
characteristic of this light per- 
mits such an unorthodox pro- FIG. 11. The hand light. 
cedure to be satisfactorily carried 
out without leaving an apparent hole in the set illumination. 

Our modeling equipment is completed by the addition of two other 
units, the portable foot light (Fig. 10) and the hand light (Fig. 11). 
This floor light, working with and ahead of the close-up camera, is 
maneuvered to highlight the subject properly from this camera angle. 

50 W. C. EDDY IJ. s. M. P. E. 

Such a technic decrees that the intimate close-ups which produce the 
best delineation of halftone value shall benefit by the best lighting. 
It is impossible, of course, to light each shot of each camera from the 
optimum angle in a studio where we find the duration of pick-up from 
a single camera sometimes a matter of seconds. We have, therefore, 
made it a practice to work toward the camera that best displays our 
wares, after making sure that the foundation lighting over the set 
is so arranged as to supply satisfactory illumination for the other 

The hand light (Fig. 11) is used to reinforce floor light in such se- 
quences where a single camera shot can be safely modeled to the con- 
trast limit. It is normally used on the wide-angle close-up camera 
and can be fitted with either a spotlight for contrast highlights or a 
diffusing lamp for the more subtle modeling. 

We do not attempt to approach the contrasts common on the stage 
and in motion pictures. In television we are confronted with a highly 
compressed contrast range that permits modeling, to be sure, but also 
holds as a penalty for exaggeration a wash-out or a complete black. 
It is therefore necessary that we work well within these limits, since 
the review and criticism of our lighting technic is by the audience in 
the field rather than by a cutting-room jury. This, however, has 
not restricted the use of modeling light; the trend, on the other hand, 
being toward the greater contrast that the electrical system will ac- 
cept, in preference to the flat non-dimensional pictures of past years. 
Experience gained by operation and observation appears to be the 
only rule in the use of these modeling fixtures even though we have 
endeavored to take guesswork out of the equipment. 

Our failure to mention back-lighting does not mean that we have 
overlooked the possibilities of this type of illumination. In the studio 
sets we have yet to arrive at a reasonable system of back-lighting 
that will answer all the requirements of flexibility, weight, and op- 
eration. It is true that we now are using, in our main studios, an 
advanced type of remotely controlled ceiling light that appears to 
solve the problem, but since our findings to date are not conclusive, 
we felt that discussion of this system should be held for the future. 

We make use of one other type of light that merits consideration. 
This equipment is known as the "portrait table," used as the name 
implies: in cases where the picture is primarily a portrait. Four 
lights are arranged at the outer rim of the announcing desk on flexible 
goosenecks adjustable as to height, angle, and throw. By substitu- 


tion of various types of bulbs and variation of the wattage, detailed 
modeling of the face can be effected with a minimum of difficulty. 
This equipment also has portable back-lighting, which again is con- 
trollable, making the work shot of this table the television equivalent 
of a studio portrait. 

This enumeration completes the catalog of our lighting technics 
and equipment in the National Broadcasting Company television 
studios. We have tested all reasonable systems of light production 
and are still carrying on these investigations. Lately we have been 
interested in vapor-lamps as a possible adjunct to the system, but 
the complications inherent in a three-phase power-supply and a water- 
cooling system would appear to make further consideration of present 
models impracticable. 

There have been many statements and many more conjectures as 
to the light used in television studios. We quote pertinent figures 
based on our last six-month period of operation. Our average set illu- 
mination was in the neighborhood of 1200 foot-candles of incident 
light. Our average modeling ratio was 2 to 1, while the average light 
load was slightly more than 50 kw of 110- volt d-c. Our lowest 
foundation lighting level was 800 foot-candles, a play in which the 
contrast throughout the set was carried to the upper limit of the 
Iconoscope. The highest foot-candle reading recorded was slightly 
less than 2500 foot-candles, a continuity where, obviously, little 
modeling was attempted. 

In our work of the past three years, we feel that we have established 
a substantial foundation in television studio lighting on which we 
hope to base an even simpler system. If we appear to have stand- 
ardized certain assemblies and particular light-sources, this does not 
mean that our developmental work has ceased. It continues with 
renewed vigor as we see our experiments bearing fruit. 


MR. ROBINSON: The liquid mercury lamp of the high-pressure type seems to 
offer good possibilities for the high level of illumination required in television 
studio lighting. This lamp has been briefly described before the Society. It is 
reported that in some tests recently in Schenectady, a very useful unit was main- 
tained using three lamps in a single unit. We used it on three-phase alternating 
current and no objectionable nicker was present. 

MR. RICHARDSON: I have been wondering why the television people have not 
been using the color-photography (C.P.) lamps? They have been working with 
what we would presumably call lamps operating at a normal tungsten tempera- 

52 W. C. EDDY [j. s. M. P. E. 

ture but it seems that they might advance their technic to some degree by using 
modern "C.P." lamps, which have almost twice the photographic value. I be- 
lieve it is the purpose of the "C.P." lamp to produce a light on the side of the blue. 

MR. ROBINSON: The Don Lee Studio uses the "C.P." lamp. 

MR. CRABTREE: A question was asked at our New York meeting why it 
would not be better to photograph all of the scenes on motion picture film and 
then televise from the film thereby reducing the amount of light required as well 
as the heat. We were advised that the sensitivity of the tubes was quite equal 
to that of the photographic film. 

MR. ENGSTROM: I think one of the reasons why there is an apparent discrep- 
ancy between the sensitivity of the Iconoscope and that of present-day film is that 
recent advances have been made in film speed. A second reason involves the di- 
mensions of a film frame as compared with the dimensions of the Iconoscope photo- 
emissive mosaic. In the Iconoscope this is approximately five inches by four 
inches and the shortest lens focal length is six and one-half inches. Depth of 
field is an important limitation in direct pick-up for television. So far, increases 
in Iconoscope sensitivity have been used to permit smaller settings of lens nu- 
merical aperture and thus increase depth of field. It has been considered more 
important to have greater freedom of action in the studio than to reduce the light- 
ing. We appreciate that both the light and heat levels of today's television studio 
are too high. 

MR. LUBCKE : I might be able to satisfy Mr. Crabtree in some measure by re- 
citing an accident that took place hi our studio some months ago. The main 
lighting fuse on the d-c circuit blew about one and one-half minutes before the end 
of a "Vine Street" episode. A 500-watt indirect lamp remained lighted, being the 
ordinary 110- volt, d-c circuit. The latter part of the scene was apparent on the 
television screen as taking place in a very dense fog. 

MR. ENGSTROM: The brightness range in television images may be limited in 
several portions of the system but the present practicable limit is in the cathode- 
ray tube used for reproducing the picture. Factors that determine the limit in- 
clude the bulb-shape, the conductive coating on the inner walls, and total reflec- 
tions from the glass-air boundary. These limit the range to about 50 to 1 for 
large areas and to something considerably less than this between adjacent ele- 
ments of detail. That this is not a permanent limit is indicated by results from 
experimental tubes that have had considerably greater range between large areas 
and particularly between small details. 

MR. RICHARDSON: I think that most of us have observed that the television 
pictures appear rather flat as compared with good photographic images. In 
analyzing the writer's paper, it appears that the method described of quickly 
shifting the television camera from scene to scene all set up on a single stage pre- 
sents a serious limitation to the ultimate image as seen by the observer. Would 
it not be possible as the art progresses to establish the scenes on separate stages 
or in completely separate areas which would permit a more perfect lighting of 
each successive scene and eliminate the difficulties which television producers are 
apparently encountering when they attempt to light their scenes with a rather in- 
flexible lighting system? 

If the technic of radio studios was possible, artistic individual lighting would 
seem to be more readily accomplished. 


MR. ENGSTROM: The author of this paper has outlined experiences with the 
present limited facilities. We must remember that a production technic for 
television has yet to be developed. I believe that the methods outlined by Mr. 
Richardson will be tried. 

MR. YOUNG: What is the general opinion of the new method that is being con- 
sidered now by the Zenith Company of Chicago, that is, the direct continuity 
method of making television programs? Do you care to express yourself on that? 

MR. RICHARDSON : Probably we had better leave that for the author. 

MR. YOUNG: I have a method near enough o it that I would be glad to show 
the members so that they can see what it is like. 


Summary. The current television technical facilities of the Don Lee Broad- 
casting System in Los Angeles are briefly described. A mosaic type camera and ac- 
companying Don Lee control equipment are used. A coaxial cable conveys the signal 
therefrom to the W6XAO sight-sound television transmitters, operating on daily 
schedule on 45 and 49.75 megacycles, respectively. 

The routine of production of a dramatic comedy serial entitled, " Vine Street," in its 
forty-eighth biweekly episode at this writing, is utilized as an example. A total time 
of twenty hours of one or more members of the dramatic unit is required to prepare and 
present one fifteen-minute episode. 

The sequence of production is as follows: preparation of script; construction or 
modification of props and scenery; cast memorization of lines; cast rehearsals; 
camera, sound, sound-effects, light rehearsal with production staff; make-up; the per- 
formance itself, including visual aural introduction of the act; the performance proper 
with overall supervision of lighting, microphone, and television adjustments by a 
television-producer at a distant receiver; closing announcement; written and verbal 
report of errors or advances in technic made during the performance. 

Specifications for the physical instrumentalities and the current television technic are 
covered for each of the above factors of production. 

The television transmitter, W6XAO, of the Don Lee Broadcasting 
System, Los Angeles, went on the air on the ultra-high frequency of 
44.5 megacycles on December 23, 1931. It has been on the air with- 
out notable exception daily, except Sundays and holidays, since that 
time. In this period more than 1 1,000,000 feet of motion picture film 
have been telecast, and for over a year, experience has been gained in 
live-subject production. This paper is concerned with the latter 

A mosaic type camera and accompanying camera control equip- 
ment are located in a 25 by 50-foot studio of the Don Lee Building in 
Los Angeles. Fig. 1 shows the interior of the studio during a tele- 
vision production. The camera is seen mounted on an arm-type 
dolly, especially adapted for television by our organization. Com- 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 6, 

** Don Lee Broadcasting System, Los Angeles, Calif. 


plete panning, tilting, and elevating adjustments are available, by 
means of which it is possible to orient the camera in any position at 
elevations from l l / z to 6 feet from the floor. Focusing is accom- 
plished by a precision mechanism operated by a knob at the right 
rear of the camera case. A built-in view-finder which allows the 
image appearing on the sensitive camera tube plate to be viewed, 

FIG. 1. Don Lee television studio during a production. 

and at times a motion picture type view-finder on the side of the 
camera, are utilized for camera manipulation. 

The camera control equipment, portable along with the camera, 
is shown at the right of Fig. 2. In this equipment, voltages are sup- 
plied to the camera, and the image signal thereform is amplified and 
modified as required for the production of a television signal, com- 
plete with synchronizing and other pulses. An image monitor is also 
part of this control equipment. 



[J. S. M. P. E. 

From the television studio, which is located on the second floor of 
the Don Lee Building, a coaxial cable carries the amplified signal to 
the eighth floor, where the television transmitters proper are located. 
These are shown in Fig. 3. The transmitters comprise the ultra- 
high frequency sound unit operating on 49.75 megacycles, common 
control equipment, and the visual unit operating on 45 megacycles, 
of which only half may be seen in the figure. The sound from the 

studio is handled by the regular 
facilities of the Don Lee Broad- 
casting System, and is conveyed 
to the sound transmitter over 
sound circuits of the usual type. 
The subject matter of this 
paper will be presented by de- 
scribing the routine of the pro- 
duction of the Don Lee dramatic- 
comedy serial entitled, Vine 
Street, now in its 48th biweekly 
episode at this writing. 

The Vine Street group com- 
poses what we call a "dramatic 
unit." We found, early in our 
live-subject television work, that 
it was desirable to organize our 
various dramatic productions into 
such units. Units are organized 
in the following manner: A per- 
son or persons selected for tele- 
vision performances are inter- 
viewed, and a type of produc- 
tion decided upon which is best 
suited to the talents of the 

persons in question and also fitted to the program needs of the station 
at the time. The necessary cast is then assembled, and at least one 
writer. A week or two are utilized in the preparation of a sample 
script and in conferences with the television production department. 
A dress rehearsal is then held and needed modifications made. With 
the subsequent actual performance of the show the unit is then in 

After the unit has staged a few productions, the routine of produc- 

FIG. 2. Pick-up equipment, show- 
ing camera control unit. This unit 
is of the depressed chassis and 
panel type of construction, with 
only operating controls on the front. 

July, 1939] 



tion becomes well established, and, in general, and in the Vine Street 
group in particular, adheres to the following pattern : 

Episodes of Vine Street are presented twice weekly on Tuesday and 
Friday evenings. They are of 15-minute duration. At least two and 
often three episodes are prepared in advance. The content and ac- 
tion of the forthcoming episodes are determined by a story conference 
of cast and writers. The general theme of the serial is considered; 
the ideas, dialog, and scenes are 
evolved. The writing is done 
subsequently by Mr. Wilfred H. 
Pettitt, chief writer for the unit. 
The script is complete with re- 
spect to dialog, and camera shots 
and special effects are indicated. 
At times, and with other units, 
sketches are included of the scenes 
as they are to be taken by the 
camera. In writing, cognizance 
is taken of the fact that large and 
elaborate sets are beyond the 
present scope of television eco- 
nomically, if not otherwise, and 
that physically impossible actions 
must not be imposed upon the 

Through the use of miniatures, 
however, otherwise impossible 
action has been televised. In a 
recent episode, a considerable 
portion of the action took place in 

FIG. 3. W6XAO sight-sound 
transmitters. Each transmitter is 
crystal controlled. Only half of the 
visual unit, on the right, is shown. 

close shot with the characters in 
an airplane winging their way 

over the Pacific. Ultimately running out of gasoline, they go into 
a tail-spin and crash on land. The first scene was taken with the 
characters and life-size properties. The nose-dive was done by means 
of a miniature airplane, handled by wires, and the crash scene, previ- 
ously set up on another set, was occupied by the characters during 
the transition through the miniature. 

The characters learn only one episode at a time, and the necessary 
properties are assembled at this time or ordered if new construction 

58 H. R. LUBCKE [j. s. M. P. E. 

is involved. The principals spend several hours two days preceding 
the broadcast in memorizing their lines, and a short period going 
through the dialog together and establishing postures and action to 

The day before the broadcast, an hour is spent in refining the 
recitation of the dialog. That evening, the complete staff rehearsal 
is held. This includes the camera and lighting crew, the sound 
supervisor and sound-effects man, under the direction of our tele- 
vision producer, Charles Penman, who is also the production manager 
of the Don Lee Broadcasting System. 

The first activity at this meeting is the distribution of scripts 
to all concerned and a resume of the important camera shots and 
actions, special lighting, and sound effects, as recommended by the 
cast. These recommendations are modified by any of the several 
staff members as required. Necessary changes are agreed upon 
and the important aspects of the episode tested under television trans- 
mission conditions. Following this, a dress rehearsal is held to famil- 
iarize the operative production staff with the action. It is in this 
portion of the production that the greater part of what appears on 
the television screen is formulated. The appearance of the proper- 
ties is checked on the television monitor screen and the lighting 
arranged on the properties modified, until a satisfactory delineation 
is secured. The lighting and positioning of the cast and the camera 
angles are also determined. Microphone positions are established 
and sound effects tested. 

We have found that real properties invariably televise satisfac- 
torily, although suitable illumination may be required for emphasis. 
In painted properties, such as background, windows, and fireplaces, 
the delineation of the object from the general tone of the background 
should be sharp, and the width of lines comprising the structures 
bold. This is shown by the character of the door in the shipboard 
scene of Fig. 4. A certain amount of defocusing is usually obtained 
on the background, often for the purpose of centering attention on 
the principal characters, who are in sharp focus. The background 
properties are therefore televised in subdued tones as desired. 

For multi-character scenes, the long shot is often used with com- 
plete settings, such as a room, which may assist in the story. If 
small items of interest are to be displayed, however, the scene may 
be modified from what would normally be a long shot to one showing 
only half or two-thirds of the principals involved. One scene may 


be changed into the other by moving the camera, or by moving the 
principals. A park bench scene used in one of the early episodes 
is shown in Fig. 5. Artificial grass and a subdued woodland back- 
ground were utilized. This background must be within three feet of 
the principals in order to televise sharply. On many scenes a rather 
high camera is utilized, that is, with the lens 4 or 5 feet from the 
floor. A notable exception to this was a camera shot for an up-in-the- 
tree sequence, where the principals were supposed to be broadcasting 
a football game from the vantage point of height. Here the lowest 
camera position possible, coupled with the considerable elevation of 
a property tree, gave the desired effect of persons quite high off the 

FIG. 4. A shipboard scene from Vine Street. The 
background is painted in sharp strokes. 

ground. Changes from long shot to close-up may be made once or 
twice during an episode. Changes of scene are usually accom- 
plished by panning, under which conditions two sets are established 
on opposite sides of the general stage area. 

The technic of lighting for television appears to be one of the most 
fruitful in creating pleasing artistic effects. So-called "flat lighting" 
will give television pictures, but ones having little interest and 
sparkle compared to those televised with more elaborate lighting. 
By flat lighting is meant, of course, that nearly all the light to illumi- 
nate the scene comes from the front of the set and perhaps also from 
the top of the set at the front. 

The advanced technic appears to be limited only by the number 

60 H. R. LUBC^E [J. S. M. P. E. 

of lighting units available, and the possibility of maneuvering them as 
required for the changing conditions brought about by motion of the 
performers on the set. This problem is complicated by the fact that 
in television, illumination must be continuous for the total duration 
of the act. In motion picture technic, each portion of action may be 
made as a separate take and ample time allowed for skillful placement 
of the lights. 

In our studio, a portable switching panel is installed that provides 
control of individual or limited groups of all the lights utilized. With 
this device the lighting supervisor can vary the lighting considerably 
without touching any unit. This control is usually supplemented 
by changing diffusers, changing the angle of the unit, or by change of 

FIG. 5. Park bench scene: artificial grass, painted 
back-drop, real bench and actors. Note the dark eye 
shadow on the man. 

position of mobile units by lighting assistants. A considerable num- 
ber of the lighting units are fixed in position near the ceiling, each in 
the proper direction for usual action as has been determined by ex- 
perience. A few mobile floor units are utilized. 

Hard back lighting has feen found to be a very desirable com- 
ponent in the lighting pattern. This must be supplied by lens-re- 
flector units of the type of the MR-210. General lighting is properly 
supplied by lamps in dull finish reflectors, and modeling lights for the 
face must be diffused with one or more diffusing screens. 

The camera photoelectric tube suffers a form of overload similar to 
overexposure, if the illumination on the subjects is too great. This 
usually occurs first on the faces of the performers, giving a "washed- 


out" effect, in which the sharpness of the features is lost. This con- 
dition is eliminated by either reducing the amount or hardness of the 
light, or stopping down the lens aperture. Make-up also is a factor 
in this effect, and lighting, camera aperture, and make-up must be 
correlated in order to achieve desirable results. It has further been 
found that the spectral characteristic of the light exercises an im- 
portant effect on the resulting image. A pure white light is the ideal. 

The microphoning for an episode is determined by the action that 
takes place. Two methods of microphoning have been evolved, first 
the boom or moving microphone method, wherein a comparatively 
light microphone boom is utilized and moved to keep the micro- 
phone reasonably close to the performers. Such a boom is shown in 
Fig. 2. The usual microphone position is overhead and in front of 
the performers and as close to them as possible without appearing in 
the picture. 

The second method of microphoning utilizes up to four stationary 
microphones. These are arranged at strategic points on the scene 
of action, and the change-over from microphone to microphone is 
accomplished by fader operation by the sound monitor superviser. 
This method does not require production assistants for moving the 
microphone boom. 

Four microphones placed according to this technic may be seen in 
Fig= 1. One is above background, the second above foreground, the 
third is on a floor stand, and the fourth in front of the camera. All 
are the Western Electric type 61 8 A on this set. The floor-stand 
microphone is also a property in this scene. All modern microphones 
have been tested in our work. The type used in any performance 
depends upon the performance, and changes from time to time as the 
developmental work proceeds. 

On the night of the broadcast, the principals arrive an hour or two 
before the time scheduled for the episode. Last-minute modifica- 
tions and confirmations are made with the operating staff and new 
recommendations from the staff are received, if required. The pro- 
duction department in cooperation with the stage manager ascertains 
how the properties and scenery should be handled with respect to the 
rest of the television program of the evening. Usually Vine Street 
is the last act on the program. The scenery and properties are often 
placed on a set prior to the broadcast, and this set not used for other 

The cast is next made up. Make-up is most important in long 

62 H. R. LUBCKE [j. s. M. P. E. 

shots. In close-ups street make-up is sufficient, although accentu- 
ated make-up may be utilized by increasing the light intensity. An 
example of no make-up is given in Fig. 6. This is a photograph 
made on a receiver 20 miles from the transmitter. A ten-second ex- 
posure was utilized, during which time inescapable movement of the 
subject may have dimmed the photograph. 

A base paint approximately No. 29 panchromatic is utilized as a 
start. Eyebrows are accentuated with black or dark brown liner. 
Artificial eyelashes and eyeshadow are used. Special Max Factor 
lipstick of a brownish- violet shade is applied. This color has been 
found more desirable then the red, because the camera tube exhibits 
increased sensitivity to the red and also because red light energy is 

FIG. 6. Photograph made on a re- 
ceiver 20 miles from the transmitter. No 
make-up used; camera exposure, 10 sec. 

particularly predominant in the incandescent illumination utilized. 
Visual and aural introduction to the episode are provided by means 
of theme music, miniature stage, and appropriate introductory 
paragraph prepared by the writer of the script. In motion picture 
title fashion, a miniature stage starts the performance by the raising 
of the main curtain, the draping of a side curtain, and the retraction 
of side wings, displaying a sign reading, " Vine Street, by W. H. Pet- 
titt." The side wings are then moved to obscure the sign, which is 
immediately replaced by a second sign reading, "Starring Shirley 
Thomas as Sandra Bush." In the same manner a photograph of the 
star is next displayed, and then a sign reading, "and John Barkley as 
Michael Roberts," followed by a photograph of Mr. Barkley. Si- 


multaneously with this visual action, an off-stage announcer ties 
the forthcoming episode to the previous action and introduces the 

Cues for visual and aural production are given by arm and hand 
signals by the television producer. The camera then pans to the 
scene and the action starts. At substantially every episode that 
has been telecast, the action is continued to the conclusion of the 
episode without a break that could have been noted by the usual 
looker. Occasional lighting of subject or camera irregularities have 
required modifications of what was telecast, as compared to what was 
rehearsed, but these have been undetectable except to those in- 
timately concerned with the production of the serial. A prompting 
system has been arranged, but it is almost never called upon. 

It has been found that excellent overall supervision of all the proc- 
esses of television operation and production can be exercised by a 
suitably trained director, who observes the program at a sight-sound 
television receiver located at a representative point in the service area 
of the television station. He talks by conference-circuit telephone to 
the television studio supervisor, television transmitter control oper- 
ator, and possibly to other members of the operating staff. Defects 
in lighting, camera technic, microphoning or television control, or 
transmitter adjustments are instantly apparent to this director. 
Constructive criticisms are made to the person involved and con- 
ditions that can be cured are speedily adjusted. Monitors are pro- 
vided in the studio and also at the transmitter, the latter operating 
by radiation from the transmitting antenna. Satisfactory moni- 
toring of the performance can be achieved by the use of these moni- 
tors alone; however, the typical audience reaction secured at a dis- 
tant receiver under home-receiving conditions, the effects of slight 
interference, and other practicalities entering into the picture as it is 
unfolded on the screens of the many lookers, are all present on the 
screen of the distant director, who is usually the writer. 

On cue, the closing fade-out of the episode is made, and the camera 
switched to a sign reading, Vine Street. The fade is made electri- 
cally, and this means is usually utilized also in changing from scene 
to scene. Fades may be made as long or as short as desired. Other 
fades are used, known as out-of -focus fades and as lighting fades. 
The former is accomplished by rapidly turning the camera focus 
control so far out of focus that the scene becomes a blur of light and 
then reversing the process in coming into the succeeding scene. The 

64 H. R. LUBCKE [J. S. M. P. E. 

latter is accomplished by dimming or extinguishing the lighting units 
used to illuminate the scene. 

Coincidentally with exhibiting the sign, the theme music for Vine 
Street fades in for a short interval at full volume, then fades down, 
while the announcer gives the closing comments on the episode. At 
the conclusion thereof, the music is again raised to full volume for a 
few more bars. During all this activity, sets may be changed for any 
act that may follow. 

Reports on the pattern configuration of the lighting units, micro - 
phoning, or other production factors may be requested at this point 
by the distant director. Concluding observations or suggestions may 
also be made. Following the complete broadcast, a written report 
is prepared by the director. This includes tabulation of various 
technical readings, and the artistic observations on the merit of the 
camera shots and lighting. The body of the report consists, however, 
of one or two paragraphs summarizing the merit of the broadcast as 
a whole; procedures are formulated to prevent errors that may 
have been made by the operating staff. The report concludes with 
definite instructions to members of the staff involved, concerning 
constructional or operational changes that are to be made prior to the 
next broadcast and are to be subsequently tested during the next 
broadcast. This report is read by all concerned. It constitutes a 
running record of the television activities and has proved invaluable 
for correlating activities and for reference to the operational as- 
pects of past performances. The sheets are retained and bound once 
a year. In combination with the transmitting log, the written an- 
nouncements, and the scripts utilized in the performances, a complete 
record of television operations is obtained. 

The author is glad to acknowledge the loyal and effective work of 
Mr. W. E. Thorp, Mr. W. S. Klein, Mr. H. W. Jury, and Mr. R. L. 
Pitzer of the television technical staff; Mr. C. Penman, Mr. D. Con- 
frey, Mr. P. Faux, Mr. K. Simon, Mr. J. Peoples, Mr. W. Walde- 
grave, Mr. H. Billheimer, Mr. D. Crandal, Mr. A. Haberman, and 
Mr. H. Huber of the operative staff; Mr. T. C. Sawyer, Mr. R. 
Williams, and Mr. F. Bingham of the announcing staff; Miss W. 
Urdahl for script make-up ; and the writing and acting staffs of Vine 
Street, The Gibbons Family, The Tele-Theatre Guild, The University of 
Southern California on Parade, Dramas of Youth, The Rainbow Review, 
Betty Jane Rhodes, The Singing Strings, The Tico Tico Trio, The Sing- 
ing Chimes, Jean Markel, Fashions, Norma Young, Happy Homes, 


and numerous individuals, who, with the above, form the regular 
entertainment staff of W6XAO. 


MR. LOWNER: May I ask if the backgrounds are painted in formal black and 
white or are they in color? 

MR. LUBCKE: They can be either. We usually use black and white and 
shades of gray in which the ship scene was painted. The forest scene was painted 
in color. The color rendition is approximately the same as that of the present- 
day panchromatic film but with an unusual accentuation of red. 

MR. LOWNER: You mentioned in your paper that you preferred a white light 
source. Do you mean that the yellow or the reddish tendency of normal daylight 
tends to be detrimental? 

MR. LUBCKE: The yellowness and more especially the redness of the low- 
temperature tungsten lamps aggravates the camera's red spectral response; 
consequently the high-temperature, white motion picture type lamps are 
the preferred incandescent sources. We have experimented with and used 
mercury-vapor lamps and others particularly rich in blue. A certain amount of 
color correction can be achieved in the overall system by such methods. There 
appears to be an actual loss of detail with too much red light. With a white 
incandescent source or something toward blue the clarity is better. 

MR. Ross: Since it is unusually receptive to red, would it not be possible to 
obtain color shading by merely using different depths of red; in other words, 
different tones of red, 

MR. LUBCKE: Yes; however, red and its derivatives as a whole come out as 
white. For a time in the make-up department we utilized red lining for high- 
lights; however, because the actors looked grotesque and this reacted unfavor- 
ably emotionally upon them, we discontinued this practice and now use a white 
or cream color for highlighting the faces. 

From practical experience, I would say that we would not all want to use shades 
of red as a means of getting different shades of tone. We much prefer to work in 
black, white, and shades of gray, staying away from a very bright white. At 
times a very brightly illuminated white sheet of paper or similar thing will over- 
load the television camera tube and even the transmitter or amplifier. Last 
night, however, one of our acts was portrait sketching by an artist who, at the 
conclusion of his act, held up his sketch, which was in black and white on a per- 
fectly white paper. It was very well televised. 

MR. TREMAINE: Do you use a-c or d-c for lighting your sets? 

MR. LUBCKE: We use d-c. We have used a-c on usual motion picture type 
incandescent lamps, and have not found a discernible hum pattern on the tele- 
vision screen. With mercury- vapor lamps on single-phase a-c there is a pro- 
nounced dark and light band effect, however. 



Summary. A discussion is presented of the present-day television standards as 
adopted by the Radio Manufacturers Association, stressing their importance in rela- 
tion to the design and production of television receiving equipment. 

Emphasis is placed upon the limitations of the standards and the possibility 
of their obsolescence in the future. It is claimed that the lack of flexibility embodied 
therein is very likely to cause serious difficulties for the industry in the future, for the 
transmission of a video signal is a much more complex problem than the transmission 
of an audio signal as in modern radio broadcasting where standards were relatively 

As an aid to circumventing many of the limitations of the present tentative standards, 
a new system of television transmission standards is proposed which will greatly 
simplify the considerations involved in the design of television receivers. 

The fundamental consideration for a system for transmitting mo- 
tion pictures by radio demands in principle a system for the conversion 
of lights and shadows into electrical impulses, the transmission of these 
impulses over a distance, and their reconversion into lights and 
shadows with a minimum amount of distortion caused by the opera- 

To accomplish this objective, standards must be set up defining the 
method of operation. These standards must be sufficiently definitive 
to provide satisfactory reproduction of the intelligence at the receiving 
end consistent with contemporary engineering advances. They must 
be flexible enough that their adoption will not prevent future im- 
provements from being made in the art because of their rigidity and 
the consequent obsolescence of existing equipment and investments 
based upon their restrictions. Finally, the standards must be capable 
of providing a system for transmitting the intelligence at minimum 
cost and with minimum disturbance to other radio services. The 
last requirement demands that the radio-frequency band-width of 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
14, 1939. 

** Allen B. DuMont Company, Passaic, N. J. 



the signal be as narrow as possible in order that the signal may be 
transmitted as efficiently as possible and that there may be as many 
stations as possible in a given transmitting band. 

The present standards that have been set up by the Radio Manu- 
facturer's Association provide for the transmission of two separate 
carrier signals, one carrying the video signal, and the other the audio 
signal. The frequency-separation of the two carriers is specified in 
order that single-control tuning to any transmitter may be accom- 

These RMA standards provide for the transmission of a video signal 
of negative polarity composed of 441 lines per frame, with 60 inter- 

FIG. 1. Television film projector, DuMont statio'n 

laced fields and thirty frames per second. The television signal trans- 
mitted in accordance with these standards occupies a transmission 
band of approximately six megacycles. Synchronizing pulses are 
transmitted for controlling the sweep circuits that deflect the electron 
beam to generate the raster that develops the picture. 

The receiver designed for such a transmission system must have 
circuits designed for this type of signal alone. Any variation from 
the standards will destroy the picture, and if any such variations 
were to be adopted in the future they would cause the obsolescence of 
all existing receiving equipment. 

Engineering progress comes only through constant, continued re- 
search and development. It is obvious, therefore, that at some future 

68 A. B. DuMoNT [j. s. M. P. E. 

date our present-day so-called high-definition television will compare 
only with the crystal detector, head-phone days of radio. At that 
time, however, there will be a tremendous investment by the public 
in television receiving equipment, and the obsolescence of such an in- 
vestment will not be very cordially received. It is necessary, however, 
that such changes take place, and the problem of the present-day 
engineer is to design his receiving equipment in anticipation of such 

It is impossible for any man to predict the future, but it is possible 
for the engineering profession to exert every effort to eliminate, if 

FIG. 2. Film pick-up panel. 

possible, or to reduce to a minimum, every future trouble that may 
conceivably occur. When this reasoning is applied to our present 
set of tentative television signal standards, they fall far short of this 

All advances in television receiver design will be directed toward 
increased picture detail. As the result of engineering activity, it is 
quite possible that within the next few years developments in inter- 
mediate-frequency amplifier design, and in video amplifier design will 
readily permit the economical reception of pictures having detail cor- 
responding to an 800-line picture. With standards adopted, how- 
ever, for a 441 -line picture, and with every television receiver on the 
market equipped with synchronizing and deflection circuits capable of 
operating at only these scanning frequencies, little advantage can 
be made of engineering progress. 

July, 1939] 


The problem that arises is slightly different from the one that ex- 
isted when radio broadcasting began. The system of transmission 
was obvious; and while different systems have been proposed and 
higher-quality transmission could be employed with the adoption of 
such proposals, the fact remains that equipment manufactured twenty 
years ago is still capable of producing intelligible results from present- 
day transmissions, and this has occurred despite the fact that the 
fidelity of these transmissions is 
vastly superior to that when radio 
broadcasting first began. 

The situation is further re- 
lieved, in regard to broadcasting, 
with respect to the maximum al- 
located transmission -band width. 
The standards that were adopted 
twenty years ago provided for a 
ten-kilocycle separation of station 
carriers permitting a maximum 
modulation frequency of five kilo- 
cycles. The present policy of the 
Federal Communications Com- 
mission has been to assign carrier 
frequencies in a given area on 
widely separated channels. The 
tendency is to provide a limited 
number of stations serving a cer- 
tain area at high signal level, and to the exclusion of more distant 
transmitters. Under this policy, the five-kilocycle band- width 
limitation fails to hold, and modern broadcasters continually provide 
program service of a much higher quality. 

If a modern receiver is capable of reproducing the high-fidelity 
transmissions radiated by present-day broadcasters, the full ad- 
vantage of years of engineering is obtained by the listener. But this 
service is not gained at the expense of those who have large invest- 
ments in radio receiving equipment purchased years ago. In spite 
of the high-quality transmissions now radiated in contrast to the 
distorted signals for which these receivers were designed, the receivers 
still function satisfactorily. 

Any television transmission must necessarily occupy a wide fre- 
quency band in comparison with standard broadcast transmissions; 

FIG . 3 . Synchronizing generator panel . 


A. B. DuMoNT 

[J. S. M. P. E. 

FIG. 4. Sound and picture 
radio transmitters. 

and the number of available channels is therefore limited. With 
the exception of some unforeseen and radically new method for the 

transmission of pictures by radio, future 
development must go forward along the 
lines of improving the transmission within 
the existing channel widths, and the 
present broadcast policy of providing 
high fidelity by widely separating the 
stations in a small area can not be fol- 
lowed. Development must, therefore, 
proceed along the lines of improving 
transmission and reproduction within the 
available channels. 

This can not be the case with the 
present proposed television transmission 
system. This system rigidly defines line 
frequency, frame frequency, and interlace 
ratio, which are the only factors affecting 
picture definition, with the exception of band- width which must be 
defined to give the system any semblance of order. This proposed 
system further provides for the transmis- 
sion of synchronizing pulses to control 
the frequency of oscillation of the hori- 
zontal and vertical oscillators, and except 
for ten per cent of the time during which 
the beam is blanked out, these oscillators 
are free running. During the time they 
are free running, difficulties are ex- 
perienced in obtaining both horizontal 
and vertical resolution of the picture. 

A system of transmission has been pro- 
posed by Allen B. DuMont Laboratories 
that effectively removes many of the 
limitations of the present RMA stand- 
ards. It is definitive enough to permit 
commercial television broadcasting to 
commence at the present time, yet it 
is flexible enough to permit engineering progress to advance. This 
is accomplished, in brief, by defining carrier separations and band 
width, which must be standardized, but it permits the television 

FIG. 5. Fourteen-inch Du- 
Mont console receiver. 

July, 1939] 



signal to be transmitted with any number of lines and frame fre- 
quency, and any interlace ratio. 

By dispensing with the sweep oscillators ordinarily required in the 
television receiver designed for reproduction of RMA standard trans- 
missions, the problems of properly designing the deflection circuits 
are greatly simplified. They are no longer required to match ac- 
curately the operation of the deflection circuits of the transmitter 
while running free during ninety per cent of their operating cycle. In 
lieu of this type of operation, this new system provides for actual 
transmission of the deflection voltages during their entire operating 

FIG. 6. Comparison of received picture and original film. 

If a receiver be properly designed for operation from the DuMont 
system of transmission, the deflection circuits consist only of ampli- 
fiers having the requisite gain and voltage output to generate any 
type of voltage-wave shape that may be employed to develop the 
television picture. With such a system of transmission, the deflec- 
tion of every receiver operating from a given transmitter is under the 
control of the circuits at the transmitter for 1*00 per cent of the 
operating cycle. Because of this control, the position of the beam 
at the receiver can not deviate from the position of that at the trans- 
mitter and resolution problems due to such inaccuracies are prac- 
tically eliminated. 

With the electron beam at the receiver under complete control of 
the transmitter, it can be operated upon in any manner desired. In 
areas where the income from the service does not justify the invest- 
ment necessary in a high-definition television transmitter, or in cases 
where, for some other reason, definition corresponding to the RMA 

72 A. B. DuMoNT [j. s. M. P. E. 

441-line transmission is not required, the transmitter may radiate a 
sweep signal of lower frequency, and it will be readily reproduced by 
a standard, readily available, commercial receiver designed in ac- 
cordance with the proposed DuMont system. This same receiver 
may be used for reproduction of transmissions having 441 -line detail. 
In the light of future developments, this same television receiver, 
designed in accordance with the proposed standard, will still be ca- 
pable of reproducing the television signal transmitted in accordance 

FIG. 7. Received picture from Paramount newsreel 
over DuMont transmitter. 

with such future developments. The interlace ratio may be controlled 
accurately at the transmitter to prevent the pairing of lines found in 
the present system when the sweep circuits are free running. The in- 
terlace ratio and the line and frame frequencies are determined at the 
transmitter and may be varied to provide any transmission detail de- 

Plans for design and production .of television receivers are under- 
taken, at the present time, with certain misgivings in regard to what 
may be expected in the 'future. The design of equipment in accord- 
ance with such a rigid set of specifications as are proposed for the 
RMA system of transmission must be unnecessarily detailed ; and the 
production manufacture of such equipment will be much more costly 
than might be anticipated by drawing a comparison between the 


modern television receiver and the modern radio receiving set. Fur- 
ther, the manufacturer of such television receiving equipment may be 
justifiedly worried that his reputation will be destroyed by the com- 
plete obsolescence of such equipment after it has been marketed. 

At the present time, the fundamental design consideration of a 
television receiver is, "Will it work five years from now?" With this 
thought in mind, the DuMont television system has been proposed. 
It is believed that this television transmission system offers much to 
the engineering profession in establishing a balance between rigid, 
restricting, fundamental definition and the maximum possible flexi- 
bility consistent with providing a service to the public. 


MR. LUBCKE: In 1936 we were utilizing what is generally known as the usual 
television system, in contradistinction to the DuMont television system. In one 
of our demonstrations we found that the number of lines transmitted by the 
transmitter had accidentally changed by five per cent during the demonstration. 
We at the receiver did not know of it until after the demonstration, when I was in- 
formed of it by the transmission operator. The point is that the usual synchro- 
nizing-pulse type of television system in rather general use throughout the world is 
capable of "hanging on" at the receiver to changes that even spasmodically occur 
at the transmitter. With slight readjustment at the receiver, perhaps a ten or 
fifteen per cent change could have been accommodated; and with a visit from a 
competent service man, perhaps as much as fifty or one hundred per cent. 

MR. RYDER: There has been a lack of information here with respect to large- 
screen televising, although some of us are aware of the fact that large-screen tele- 
vising is being done in England. We would welcome more information with re- 
spect to large-screen television. The picture in normal "so-called" instantane- 
ous television is actually out of synchronism with the sound. This results from 
the picture-image storage in either or both the camera and receiving tube. The 
delay is about equivalent to a frame and a half of motion picture film as observed 
by those of us familiar with "out-of -synchronism" in terms of motion pictures. 
An interesting fact in this regard is that in television the picture lags behind the 
sound, whereas in ordinary speech the sound lags behind the picture or the image. 
This has great importance as we consider large-screen picture projection. We are 
quite aware of the fact that in small-screen projection this "out-of -synchronism" 
condition will not cause annoyance, but it is one of the problems that should be 
given serious consideration with respect to instantaneous televising of picture and 

MR. ENGSTROM: The only system on which I have enough information to com- 
ment is the cathode-ray optical projection method. The Baird equipment is of 
this type. A small but bright image is produced on a cathode-ray tube and then 
enlarged by optical projection onto a screen. 

I would judge the performance of such a system in the present state of the art, as 
being very roughly equivalent to a not-too-good 16-mm, film projected onto a 

74 A. B. 

large screen at somewhat too low a screen brightness. Performance might be 
considered adequate because of novelty (television) or because of timeliness. I 
think the method and performance must be improved before permanent interest 
could be maintained in the theater field. And I am confident that advances will 
be made and that it is just a matter of time to work out the various problems. 

MEMBER: What are the possibilities of either eliminating the curved surface of 
the cathode-ray tube, with its distortion, or of eliminating the introduction of 
complementary distortion by optical projection. Television is young and it 
will take many years to perfect it ; many of the defects we see at present certainly 
will be overcome in the future. 

MR. GREEN: Apropos of that distortion I think Mr. Engstrom's company now 
has a tube which will overcome some of this trouble. In this tube I think the 
ratio between the dark point and the light point together exceed 102. That is 
one way of eliminating the curve effect. Another way that seems fairly obvious 
and which I think has been done would be to project the cathode-ray beam on to a 
phosphorescent screen, a flat screen, and look at it. There is a window in the 
tube. Of course, you get distortion of somewhat the same type as if you took a 
movie projector and shot it up at an angle. I just wondered if this line had been 
carried very far. If we see the image on the side where the light is being generated 
it is always clearer than it would be if laid on the back side which is what we 
would do with ordinary television. I do not know whether this is being carried 
forward or not but it will certainly eliminate the effect that is being objected to. 

MR. JOY : In England the Baird and the Scophony systems are being used for 
projecting television images on large screens. Is the Scophony system being used 
in this country? If so, who is experimenting with it? 

MR. WILLIFORD: It is being investigated but I do not know what has been done 
about it. 

MR. SMITH: A certain amount of confusion is permissible in transmission in 
comparison with that required by moving picture work. Television, apparently, 
can stand a good deal more. I wonder if anyone has any figures on that. 

MR. GREEN: Assuming that your lens were of such high quality that you 
could disregard the minute size of the circle of confusion, what would then be the 
limiting factor, or what is the size of the scanning dot ? That in turn is found in 
the number of lines which is 300 out here* and 441 in the East. We can also 
take into consideration the size of the screen and television camera. I do not 
know what it is exactly but I believe it is four and a half inches by six inches. 
On that basis the circle of confusion would probably be large with regard to 
photographic standards. However, it seems to be good enough for the present. 

* Changed to 441 lines, May 15, 1939. 


Summary. A report on the aims and work of the Committee, with partial 
reports by the two sub-committees: (A) on Television Production and Reproduction 
Technic, 0. B. Hanson, Chairman, and (5) Film Properties and Laboratory Prac- 
tice, 0. Sandvik, Chairman. The scopes of activity of the sub-committees are de- 
scribed, and their program for the coming year. Among the items covered by these 
scopes are (1} glossary, (2) bibliography, (3) tutorial material, (4} dimensional 
practices, (5) normal equipment and procedure, (6} special problems such as inter- 
industry coordination, future equipment needs, and specifications, etc. 

On March 9, 1938, the organization meeting of the Television 
Committee was held at the Hotel Pennsylvania, New York, N. Y. 
Since that time, considerable study has been given to the range of 
activity of the Committee, its scope of interests, and the manner in 
which its work is to be conducted. Although a limited number of 
meetings have been held, aside from the organization meeting and 
the meetings of the Sub-Committee Chairmen, much preparatory 
work has been done by conference and correspondence. The result 
of this work is not forthcoming at this time in specific form, because of 
several reasons: first, it is necessary that the Committee and its 
Sub-Committees adjust their viewpoints with respect to the relation 
between the motion picture industry and the television art; second, 
most of the subjects engaging the attention of the Committee are 
long-term projects; and, third, the television art is in so marked a 
state of flux at the present time that great care must be taken to differ- 
entiate between accepted practice and transient developmental 

In view of these facts, the present report of the Committee will be 
concerned mainly with its organization and scope of activities, with 
the expectation that by the time of the Fall Convention of the Society, 
the Committee will be able to report more specifically on at least some 
of the projects described below. In reporting on the state of the 
art, it is the purpose of the Committee to avoid causing undue or un- 
justified concern to the motion picture industry or giving inaccurate 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
4, 1939. 



ideas as to the imminence of large-scale commercial television de- 
velopments or the mode of utilization of the products of the motion 
picture industry. As a matter of policy, only factual (scientific and 
technical) matter is to be included in these reports, opinions being 
avoided so far as possible; and no attempt is to be made at any time 
to issue either reassuring or alarming statements or non- technical 
generalities. In any statements made by the Committee, at least 
for the present it should be definitely understood that these reports 
do not represent any ideas of permanence or finality in the practices 
of standards discussed. The reports are intended solely as informa- 
tive guides for motion picture and television engineers. 

Scope of Activity. The Committee will endeavor to collect, formu- 
late, clarify, and disseminate useful information to the motion picture 
industry as to television film and pictorial requirements, and to the 
radio television groups as to motion picture capabilities and avail- 
ability. It is hoped to avoid conflicting standards or practices in the 
two arts. The membership of the Committee includes prominent 
members of both industries so that an automatic liaison will exist 
without the necessity of delegating any particular agency or person 
for the purpose. 

Reports will be made at timely intervals, as developments may 
direct, by the Committee as a whole or by the Sub-Committees. 
The first aim of these reports is to be historical and instructional, 
and, accordingly, to collect and collate existing material. The second 
object is to guard against misunderstanding misstatements in the 
press, unnecessary conflicts of aims or opinions, and to obviate or 
reconcile these wherever possible. The third purpose is to act as one 
guiding agency in directing technical activities common to the two 
industries and furthering interchange of mutually helpful data. 

Two Sub-Committees have been established Sub-Committee A 
on Production and Reproduction Technic, under the Chairmanship 
of O. B. Hanson; and Sub-Committee B on Film Properties and Lab- 
oratory Practice, under the Chairmanship of O. Sandvik. 

Considerable difficulty was experienced at first in outlining the 
scopes of the Sub-Committees so that they would cover the necessary 
phases of the art and permit a coordinated course of action, without 
overlapping and duplication of work. It is impracticable at the pres- 
ent time to report specifically on a number of the items falling within 
these scopes, due to the fact that the technic of television is in a state 
of flux and many phases of the art are really in the experimental stage. 


The characteristics and sensitivity of Iconoscopes (pick-up image 
tubes) have been changing as new and better models were produced. 
These changes affected the lighting set-ups, the lenses, the treatment 
of the sets, and other phases of production technic, and any report 
that might be made at the present moment might be obsolete and 
misleading by the time it reaches the members of the Society. 

The Iconoscopes used for the transmission of images originating 
from motion picture film have somewhat different characteristics 
from those used in the studio. Due to the rapid changes occurring 
in the development of these Iconoscopes, the density and gamma of 
the films used could not be determined under these circumstances, but 
it was generally felt that, rather than have prints made with special 
characteristics for television requirements, it would be better at this 
stage of the art to endeavor to attempt to accept the standards of the 
motion picture industry at least until such time as the limits of the 
television system could be finally determined and stabilized which 
limits might subsequently lead to specifications for specially processed 

As a basis of a temporarily acceptable policy for the industry, the 
opinion appears to be that the present motion picture standards are 
acceptable for television and that television will try to work toward 
those standards. Motion picture standards should not be degraded 
to meet television requirements. There are differences, however, 
between the requirements of the television art and those of the motion 
picture art. For that reason, much of the early work of the Com- 
mittee is to be of an educational nature that is, the collection of 
pertinent information of interest to the motion picture industry. 
There are differences as to set construction, scenery, and limits as to 
detail, size, and coloring. There are also limitations of systems in 
relation to the sensitivity of the Iconoscope, types of light-sources, 
floor and overhead lighting, long shots and close-ups, modeling, and 
air-conditioning. The subject of lenses is closely connected with 
those of film and lighting. Some of the concepts are different in the 
two arts, for example, what is called a "dissolve" in television, is 
really what is known in motion pictures as a "fade-in" and "fade- 
out." Very little is known at the present time in a final sense with 
regard to mobile equipment and further developments must be 
awaited. Problems of background projection are similar to those for 
the motion picture except with regard to the light required. The 
Iconoscope screen size is not yet definitely standardized, and although 


projectors use standard sizes of film, they are modified to accommo- 
date the different rates and methods of projection. 

To be more specific, the proposed projects of the two Sub-Com- 
mittes have been more or less definitely divided into the following 
groups : 


Production and Reproduction Technic 

(a) Glossary. Material for the glossary will be obtained from all sources 
including the publications of the Institute of Radio Engineers, Radio Manu- 
facturers Association, Society of Motion Picture Engineers, Acoustical Society 
of America, Optical Society of America, and the like. 

(6) Bibliography. A list of articles and books dealing with television will be 
prepared so far as motion pictures, film, film photography for television, and 
film projection and transmission for television are concerned. It may prove 
advisable in some instances to include abstracts of the articles or summaries of 

(c) Tutorial. General descriptions of television equipment, methods and 
use of films, with respect to their present status and probable trends. 

(d) Dimensional practices. 

(e) Normal equipment and procedure practices. 

(/) Special problems (inter-industry coordination, future equipment needs 
and specifications, and the like.) 

With respect to the available data, items (a), (6), and (c) will be 
given preferred attention with early stress on (a) . It is hoped, how- 
ever, that eventually the entire work of the Committee will be ex- 
panded to include all the items listed above. 


Film Properties and Laboratory Practice 

(a) Glossary (as before). 
(6) Bibliography (as before). 

(c) Tutorial. A general description of television film equipment in relation 
to photographic requirements, desirable film characteristics, and exposure and 
processing conditions of negatives and prints. 

(d) Standards. Dimensional practices as related to the use of motion picture 
films and film handling equipment. 

(e) Normal equipment and procedure practices. 

It is recognized that a considerable amount of work will be re- 
quired by the Sub-Committees to gather the material for the survey 
and prepare the glossary, as well as to undertake work on the other 
phases outlined above. It is felt that nomenclature and the glossary 

July, 1939] 



provide the logical starting point for the work, which should be fol- 
lowed immediately by the bibliography. Accordingly, it is hoped 
that at the Fall Convention of the Society, the Committee may re- 
port on these subjects, and, in addition, it is felt that sufficient ex- 
perience may have been gained by that time to report on some of 
the other projects. For example, a limited television service will 
be available this season in the New York metropolitan area from one 
or more of the television broadcasters. The experience gained in 
actual operation during 1939 may assist materially to stabilize pro- 
duction technic, or at least permit the preparation of a report cover- 
ing some of the phases of production. Some time will probably pass 
before production and reproduction technic of television will reach a 
stage of stability such as to permit a determination of standards of 










A. B. DuMoNT 

and Reproduction Technic 
O. B. HANSON, Chairman 

ties and Laboratory Practice 
O. SANDVIK, Chairman 



Summary. Sound reproduction systems are designed on the premise that the 
sound-track will be illuminated by a scanning-beam of substantially uniform flux 
density. This paper presents results of extensive studies of the actual beam charac- 
teristics for all types of optical systems and lamps employed in the reproduction of 
sound from film. They were made possible by a unique microphotometer, designed 
by the author, with which the scanning beam can be analyzed in minute elements. 

Th studies cover: relative levels of scanning beam illumination; effect of source 
displacement from design position on total flux at the sound-track; microphotometer 
recordings of distribution of flux density across the beam as affected by optical systems 
and source forms and by displacements of the source. 

In reproducing sound from film the uniformity of sound-track illumi- 
nation is becoming more and more a subject of interest to equipment, 
lens, and lamp manufacturers alike. Uneven illumination distorts 
tone quality and may cause volume attenuation. 1 Substantially uni- 
form illumination is possible with any of the three characteristic types 
of optical systems currently in use. Failure to achieve such uni- 
formity may be due to the design or lack of precision of the optical 
system, the use of a lamp of wrong design for the purpose, lack of pre- 
cision in the source, or inaccurate adjustment of lamp position. This 
paper summarizes the results of extensive studies of the effects of 
source form and positioning on uniformity of illumination of the 

Test Equipment. All the data presented in this paper were ob- 
tained by a recording microphotometer designed by the author. 
There had not previously been available facilities of a kind which 
made it convenient to carry out such studies of adequate extent. 
The instrument is, in effect, direct reading, measuring scanning beam 
brightness in terms of the response of a typical caesium cell. Thus 

* Presented at the 1939 Spring* Meeting at Hollywood, Calif. ; received 
April 13, 1939. 

** Nela Park Eng. Dept., General Electric Co., Cleveland, Ohio. 



the complications attendant upon the use of standard test-films are 

The microphotometer is shown in Fig. 1 and schematically illus- 
trated in Fig. 2. The essential elements include : 

(7) A mounting for the exciter lamp permitting adjustment of source position 
in three dimensions in steps of 0.001 inch, and means for tilting the source by 
known amounts in the vertical plane. 

(2) A mounting for the sound-reproducing optical system permitting the 
scanning beam to be focused in the plane of the film. 

(5) A slit in this plane less than 0.001 inch wide in the direction of the long 
dimension of the scanning beam. Thus, with the slit in place, the flux in each 

FIG. 1. The recording microphotometer. 

element of less than 0.001 inch of the scanning beam may be measured or, by 
removing the slit from the beam, the total flux. An additional adjustable slit 
may be substituted and set for a band of any desired width. 

(4) A small integrating sphere and caesium photoelectric cell to receive the 
flux from the scanning beam. The integrating sphere is used to eliminate errors 
due to differences in sensitivity over the cell surface. 

(5) A direct-current amplifier 2 for the photocell current, the output of which 
may be read on a microammeter or by the deflection of a high-sensitivity gal- 
vanometer when the scanning beam is being explored with a narrow slit. 

(6) A galvanometer lamp to project a beam of light onto the galvanometer 
mirror and a film drum to record the galvanometer deflections as a trace of the 
reflected beam on a moving film. 

(7) A mechanical linkage between the slit and the film drum so that the 
abscissas of the exposed traces bear a direct relation to the movement of the 

82 F. E. CARLSQN [j. s. M. P. E. 

slit, and suitable motive power to provide synchronized automatic movement 
of slit and drum. 

Types of Optical Systems. The optical systems tested are of the 
types shown in Figs. 3a, 3b, and 3c. They are in use in both 16-mm. 
and 35-mm. sound-reproducing equipment. The system illustrated 
in Fig. 3a closely resembles that employed for motion picture projec- 
tion. The light-source is imaged substantially in the plane of a me- 
chanical slit (corresponding to the aperture), which is, in turn, imaged 
on the film plane (corresponding to the screen) by an objective lens. 
The light-source is of the same general proportions as the slit, long in 
relation to diameter. Usually about 0.0015 inch of the filament coil 
diameter and about 0.126 inch of the coil length are effective. 

The system illustrated in Fig. 3b closely resembles that of a stereop- 
ticon. Here the condenser forms an image of the light-source in the 

FIG. 2. Schematic diagram of recording microphotometer. 

objective lens and the mechanical slit is placed close to the condens- 
ing lens. Since the aperture of the objective lens is circular, the useful 
portion of the light-source must lie within a circle of appropriate di- 
ameter (usually between 0.070 inch and 0.100 inch). This calls for a 
source relatively much shorter in length and much larger in diameter 
than those used with the "motion picture" type of system of Fig. 3a. 
The system illustrated in Fig. 3c usually consists entirely of cylin- 
drical lenses, although spherical surfaces are sometimes included. 
In this type of system the smaller dimension of the source is imaged 
directly on the film, and reduced in size to one-tenth or less, by the 
lens or lenses nearest the film plane, thus determining the short di- 
mension (width) of the scanning beam. Since it is not confined by a 
limiting aperture, a change in the diameter of the coil produces a cor- 
responding change in the width of the scanning beam and a variation 
in the resulting sound volume at different frequencies. 3 - 4 Similarly, 

July, 1939] 



tilt or distortion in the coil changes its width* and is reflected in a 
corresponding change in the width of the scanning beam and in the 
amount of amplitude distortion. Lamps with coils of comparatively 
small diameter are therefore used, selected within denned limits for 
source width (usually 0.013 inch). 





FIG. 3a. The "motion picture" type of optical system. 






FIG. 3b. The "stereopticon" type of optical system. 



FIG. 3c. A "cylindrical lens" type of optical system. 

In the plane of the coil axis, the cylindrical system is usually essen- 
tially of the "stereopticon" type. The condensing lens A (Fig. 3c) 
forms an image of the coil length in the objective lens B and an aper- 
ture is placed close to the condensing lens whose image defines the 
length of the scanning beam. Coil length is important only to the 
extent that it must be sufficient to cover the intermediate lens as in 

* Source width as used here is measured in a plane parallel to the film plane 
and is the separation between lines through the highest and lowest points of the 
source, both lines being parallel to the designed position of the coil axis. 



[J. S. M. P. E. 


3| g 


C C 1C C ^ 


g 2 S 2 ^O 

1 1 1 1 1 1 *- 

! IB 








^ 1C 1C O O C 

rt3v O CO CO Tj< OJ 

I I 

CO 00 CO 00 00 

^ as 5-0000000 


c o o o o 

|lS~! 8888888 

" ^coocddo 

8 88888 

d d d d d d 

a O O U 

.2 S|S8888l 


GO CO CO 03 O3 



s s 

CO C CO l> 00 CD 

c Tt rf co cq cq 

^ * u 0. ^ i 1C C O 1C 

" -= = ^ -; 5. - 

-5 S c 


-> J^ 
pa < 


C5 i-i 

<M CO 







Xi T I 1 I Tt CO 

i rt^ C O C O 

.o *? d d d d 

O Tt< 

C O O O 

E - E 

o o o 

o d d d d d 

5 co 
>"& o 

9 ^ 

d d d 

O O -i 

o o o o o 

o o 


5 C5 lO O O ^ 

o o o p o p 

00 O5 O O C t^ 

July, 1939] 



Fig. 3c, or sufficient to provide the requisite length of scanning beam, 
if imaged directly on the film. 

Types of Lamps and Their Characteristics. The data included in 
this paper are based on tests with the lamps listed in Table I, except 
in a few instances where great extremes in coil diameter or coil length 
were needed to indicate more clearly the effect of these factors. 

Sound-track illumination varies with optical systems of different 
manufacture, as well as with the form of source employed and the 
current at which the lamp is operated. Table I gives the total light 
flux found for the optical systems of one manufacturer. In obtaining 
these data three systems of each 
type were tested and the scan- 
ning beam length was masked as 

Where in subsequent charts 
uniformity of sound-track illu- 
mination is shown, the recordings 
apply for the individual system 
of each group of three which was 
most perfect in this respect. This 
is done in order to concentrate 
attention on source rather than 
on lenses. Some of the variation 
exhibited in the curves is, however, still ascribable to defects in the 
optical system. 

The hum resulting from the light modulation of the exciter lamp 
operating on alternating current is a familiar phenomenon. The 
extent of this modulation is roughly proportional to the thermal in- 
ertia of the tungsten wire and is reduced as filament diameter is in- 
creased. It is also reduced when the frequency is increased. In 
Table I are also included data on the per cent modulation of the light 
when the lamps are operated at their respective rated currents on 60- 
cycle alternating current. 

Some of the listed lamps are fitted with the bayonet-candelabra 
base, others with the newer bayonet-prefocus base. With the former, 
the accuracy of filament positioning with respect to the reference 
planes of the base is usually =*= 3 /64 inch (0.046 inch) . When the bayo- 
net-prefocus base is used these tolerances are reduced to 0.010 inch. 
Any of the lamps may be ordered so equipped. The significance of 
this more accurate base will be evident from the data to be presented 

FIG. 4. The X t Y, and Z axes indi- 
cate the directions of source displace- 
ment in the tests covered in the text 
and charts. 



[J. S. M. P. E. 

30 20 10 IO 20 



FIG. 5. The effect of source displacement along the X and Y axes on flux 
through aperture and uniformity of sound-track illumination "motion pic- 
ture" system. 

. 20 40 6O SO K5O 


FIG. 6. The effect of source displacement along the optical (or Z) axis 
on flux through aperture and uniformity of sound-track illumination 
"motion picture" system. 

July, 1939] 



on the effect of source displacement on sound-track illumination. 
For convenience in referring to source displacements they will be 
referred to as the X axis, Y axis, or Z axis, as shown in Fig. 4. The X 
and Y axes are parallel to the plane of the film. The X axis is parallel 
to, and the Y axis is perpendicular to the long dimension of the scan- 
ning beam. The optical axis is designated as Z. Except where stated 
otherwise, it may be assumed that the axis of the coiled filament co- 
incides with X. 







te=: C^ 





in ' 

' A 


V -2*L 




-4 L 





u -a 






FIG. 7. The effect of filament tilt in the X- Y plane on flux through aper- 
ture and uniformity of sound-track illumination "motion picture" system. 

"Motion Picture" Systems. The two lamps most frequently used 
with the "motion picture" system (Fig. 3a) are the familiar 8 1 /2-volt, 
4 -ampere and the newer 8- volt, 2-ampere. In most 35-mm. repro- 
ducers, it has been the practice to operate the former at 3.2 amperes, 
since at this value photocells currently available provide ample re- 
sponse. Between 4-ampere and 3. 2-ampere operation there is a 
difference in sound level of about 10 db. The newer 8- volt, 2-am- 
pere lamp produces at rated current substantially the same sound- 
track illumination as does the 8V2-volt, 4-ampere lamp at 3.2 am- 
peres. At its rated current of 2.0 amperes the service life of the lamp 
is comparable with that of the 8V2-volt, 4-ampere lamp at 3.2 am- 



[J. S. M. P. E. 

The scanning beam produced with this type of system is invariably 
considerably longer than the required 0.084 inch and therefore all 
measurements of sound-track illumination which follow are limited 
to the cell response from the middle 0.084 inch of the scanning beam 

The curves of Figs. 5 and 6 show the effect on sound-track illumina- 
tion of displacing filaments from the design source position along the 
X, Y, and Z axes. Due to the steepness of the curves for displace- 
ment along the Y axis the data for only one coil diameter are shown. 
The effect on uniformity of illumination is shown by the recordings. 
In Fig. 7 is shown the effect of filament tilt in the X-Y plane. The 
8-volt, 2-ampere lamp has been used for these recordings because of 
the more uniform illumination it produces, as shown in Fig. 8. 


FIG. 8. Recordings of sound-track illumi- 
nation using (a) 8-volt, 2-ampere lamp (solid 
curve) and (b) 8^-volt, 4-ampere lamp (dotted 
curve) "motion picture" system. 

Since, with this type of system, the source is imaged substantially 
in the plane of the slit, it is reasonable to expect that some non-uni- 
formity in sound-track illumination is inevitable due to differences in 
brightness from turn to turn in the coiled filament. Actually, such 
non-uniformity is considerably influenced by the design of the coiled 
filament as shown in Fig. 8 and also by the fact that the mechanical 
slit in the systems submittted for these tests is curved to correct for 
curvature of field by the objective lens. Probably the presence of 
spherical aberration in the condenser lens is also helpful. It will be 
noted that, with the 8V2-volt, 4-ampere lamp, there is evidence of 
periodic changes in intensity due to individual turns of the coil but 
that it occurs only where, apparently, the filament image is critically 
focused on the curved slit. Such periodic changes in brightness along 

July, 1939] 
































X 1 

IOO 80 60 40 20 20 40 60 80 



FIG. 9. The effect of source displacement along the optical or Z axis on 
ftux through aperture and uniformity of sound-track illumination with "stere- 
opticon" systems. Curve is typical for several sources listed in Table I and 
for optical systems having effective source areas of 0.070" and 0.100" 









Sr 6 





/ t 



_ fl Z 






85 o 








/ 1 





3 U 





/ , 





1 ' 










\6 -36 

FIG. 10. The effect of source displacement from design position along the 
X and Y axes on flux through aperture and uniformity of sound-track il- 
lumination "stereopticon" systems. Curves for displacements along X 
axis are typical for systems having effective source areas of 0.070" and 
0.100" diameter. Curves for displacements along the Y axis show average 
values for several sources listed in Table I. Curve A applies to a system 
having an effective source area of 0.070" diameter; curve B, to one of 
0.100" diameter. 



Q. S. M. P. E. 

the entire length of the scanning beam have been noted in previous 
tests with other systems of this type. 

The use of ribbon-filament sources for this type of system has been 
proposed from time to time. Such a source does eliminate many of 
the minor irregularities evident in the recordings, but because of its 
lower luminous efficiency, it provides a much lower level of scanning 
beam illumination per watt consumed, and invariably must be of 

FIG. 11. The effect of source displacement, from position for maximum 
illumination, along X and Y axes on flux through aperture and uniform- 
ity of sound-track illumination "stereopticon" system. 

Curves for displacements along X axis are typical for systems having 
effective source areas of 0.070" and 0.100" diameter. Curves for dis- 
placements along Y axis show average values for several sources listed in Table 
I. Curve A applies to a system having an effective source area of 
0.070"' diameter; curve B, to one of 0. 100" diameter. 

much higher wattage than desired by the designer of the power 

Referring again to Figs. 5 and 6, it is apparent that the bayonet- 
prefocus base for lamps used with this type of system obviates all 
the usual lamp adjustments except for positioning the source along 
the Y axis. This adjustment could also be eliminated through the 
use of light-sources of considerably larger coil diameter, but only if 
coil length were at the same time reduced. Such reduction is, unfor- 
tunately, accompanied by some sacrifice in uniformity of sound-track 

July, 1939] 



illumination because of the lower temperature of the ends of the coil. 

The " Stereopticon" Systems. The relative levels of illumination 
from the various sources used with the "stereopticon" type of system, 
Table I, necessarily depend upon the aperture of the objective lens 
and the magnification of the source image. Rarely is this combina- 
tion such that the objective is completely filled by the diameter of the 
coil image, although it is always filled by the length of the coil image. 

It is interesting to note that, unlike the "motion picture" type of 
system, maximum illumination is not obtained at the design position 
for the source (Fig. 9). As shown by the recordings in Fig. 9, uni- 
formity suffers when the source position is adjusted for maximum 


FIG. 12. Recordings of sound-track 
illumination using: (A) straight fila- 
ment coil on X axis, and (B) curved 
filament coil in X-Z plane "stereopti- 
con" system. 

illumination. This is particularly true if the source is displaced from 
the optical axis as shown by comparing the recordings of Figs. 10 
and 11. 

Here again, the advantage of using filament coils longer than re- 
quired merely to fill the system is evident. Moderate displacement 
of the recommended sources along the axis of the coil has relatively 
little effect on illumination. Positioning of the source in the direction 
perpendicular to the axis of the optical system continues to be the 
most critical adjustment, but less so than with the "motion picture" 
type of system. It becomes still less critical when coil diameter is 
increased materially beyond the value required to fill the objective 
lens or when coil diameter is drastically reduced so that the lens is 
only fractionally filled. 



[J. S. M. P. E. 

With this type of system, the effect of filament tilt at the design 
position for the source is negligible and therefore no curve is shown. 
This is due to the fact that, since the source is imaged in the objective 
lens, only that part of the source lying inside a circle of appropriate 
size can be utilized. This remains a constant, even when the coil 
axis is tilted 90 degrees from its usual position on the X axis. 

The non-uniformity in the recordings shown in Figs. 9 and 10 is due 
primarily to variations in brightness of the source within the angle 
intercepted by the condenser lens. These variations are character- 
istic of coiled filaments in a plane through the coil axis. Referring to 
the diagram of this type of optical system, Fig. 3b, it will be observed 

FIG. 13. Recordings of sound- 
track illumination using straight 
filament coil (a) on the X axis and 
(B) on the Y axis "stereopticon" 

that the brightness of any point in the slit is actually the brightness of 
a limited zone of the condenser lens as viewed by the objective lens, 
which is, in turn, proportional to the light intercepted by that small 
zone of the condenser. 

With the source at the design position, uniformity of sound-track 
illumination can be improved by either of two methods. One in- 
volves curving the coil slightly and viewing it from the convex side 
with the curve in the X-Z plane (Fig. 12). The other makes use of 
the fact that the light output is relatively uniform in a plane perpen- 
dicular to the coil axis, and places the coil on the Y axis (Fig. 13) in- 
stead of on the customary X axis. The curved filament has been 
used for some time in recording equipment 5 where brightness uni- 
formity is needed on the condenser side in all planes through the opti- 

July, 1939] 



cal axis. For the stereopticon type of reproducing system, brightness 
uniformity in only the X-Z plane is effective and therefore the use 
of the straight coil in the Y axis is recommended. 

Cylindrical Lens Systems. Most of the cylindrical lens systems 
currently in use are in 16-mm. sound reproducers, although they have 
been, and still are being used in 35-mm. reproducers. They have the 
advantage of being relatively much less sensitive to source displace- 
ment. Displacements of as much as 0.020 inch along the Y axis re- 

40 30 20 10 10 30 



FIG. 14. The effect of source displacement along the X and Y axes on 
flux through aperture and uniformity of sound-track illumination cylin- 
drical lens system. 

suit in practically no diminution of total flux to the sound-track. This 
is shown by the curves of Fig. 14. 

The curves for Fig. 14 were obtained from tests of a system in 
which, as in Fig. 3c, an aperture is provided at the lens nearest the 
light-source whose width along the X axis is imaged at the film plane 
by the intermediate lens to define the length of the scanning beam. 
It is the vignetting effect of this same aperture in the Y axis that 
finally causes the abrupt loss in sound-track illumination when the 
source is displaced more than 0.025 inch along the Y axis. The effect 
of displacement along the X axis on total flux is very similar to that 
with the "stereopticon" type of system because, in this meridian, the 

94 F. E. CARLSON [j. s. M. P. E. 

cylindrical system is essentially the same. As would be expected, 
displacements along either the X or Y axis do not materially affect 
uniformity of illumination. Since this particular system is of the 
"stereopticon" type in the X-Z plane, one would expect, as in Figs. 9 
and 10, some non-uniformity hi the scanning beam due to variations 
in brightness of the source in this plane. The absence of such non- 
uniformity is attributable to the much smaller effective angle of 
light collection of the first lens hi this plane. 

No data are included on the effect of source displacement along the 
Z axis because the effect is small within the permissible limit for this 
type of system. This limit is determined by the reduction of the lens 
system in the Y-Z plane and on its aperture. These factors deter- 
mine the rate of change in the width of the scanning beam as the image 
of the width of the source moves in and out of the plane of the film. 

Conclusion. The data presented in this paper indicate the feasi- 
bility of a closer standard of uniformity of sound-track illumination 
than that covered in the 1938 recommendations of the Research 
Council of the Academy of Motion Picture Arts and Sciences. The 
results emphasize the importance of precision in optical systems and 
of accurate adjustment of lamp position. Fortunately, the need for 
service adjustment is obviated in most, if not all directions if correctly 
positioned sockets for the bayonet-prefocus bases are incorporated in 
the equipment. The data clearly show that the order of precision of 
this base is within the limits of negligible effect on sound reproduction 
limits to which many operators find it difficult to work by manual 


1 BATSEL, C. N., AND CARTWRIGHT, C. H.: "Effect of Uneven Slit Illumina- 
tion upon Distortion in Several Types of Variable Width Records," /. Soc. 
Mot. Pict. Eng., XXIX (Nov. 1937), p. 476. 

2 GOEHNER, W. R.: "Microdensitometer as a Laboratory Measuring Tool," 
/. Soc. Mot. Pict. Eng., XXIII (Dec. 1934), p. 318. 

3 COOK, E. D.: "The Aperture Effect," /. Soc. Mot. Pict. Eng., XIV (June, 
1930), p. 650. 

4 SxRYKER, N. R.: "Scanning Losses in Reproduction," J. Soc. Mot. Pict. 
Eng., XV (Nov., 1930), p. 610. 

5 DIMMICK, G. L.: "The RCA Recording System and Its Adaptation to 
Various Types of Sound-Track," J. Soc. Mot. Pict. Eng., XXIX (Sept., 1937), 

p. 258. 


MR. SKINNER: With regard to the position of the lamp, the vertical arrange- 
ment is contrary to the usual practice. 


MR. CARLSON: The vertical position for the filament is suggested only when 
used with the "stereopticon" type of system. Its purpose is to improve the 
uniformity of sound-track illumination. 

MR. SKINNER: What about the life of the lamp? 

MR. CARLSON: The life is not seriously affected. 

MR. SKINNER: In most 16-mm reproducers the practice is to control the 
volume of the machine by adjusting lamp current. I think there is a good 
chance of exceeding the limits set by the design of these lamps, and there 
certainly are more replacements than with most lamps. This is one of our prob- 
lems and I was wondering whether the lamps were designed so that we could get 
more volume out of them. 

Mr. CARLSON : There are several factors which must be considered in designing 
lamps for portable 16-mm reproducers. 

First of all, the design of portable equipments usually dictates low lamp wattage 
and current which means that the filament's diameter is small and much more 
fragile than the thicker, more rugged filaments used in lamps designed for the 
larger reproducers. 

The level of sound-track illumination, as shown in Table I, is much lower be- 
cause of the low lamp wattage. The illumination would be even lower and the 
amplifier gain would have to be increased if service lives typical of lamps for 35- 
mm reproducers are desired. In the interest of sound quality, the shorter life of 
the order of 50 hours is preferable. 

Your problem can best be solved through the use of a power supply and lamp 
of higher wattage and current, such as are used in larger reproducers. 

DR. FRAYNE: Which of the two systems, the motion picture system or the 
projection slide system, is to be preferred in case of standardization? 

MR. CARLSON: I feel that the "stereopticon" type is to be preferred. The 
system does not require so precise an adjustment of filament position as does the 
"motion picture" type to maintain level and uniformity of illumination. If the 
more accurate bayonet-prefocus base and a properly positioned socket are used 
all the manual adjustments for filament positioning can be eliminated. As 
pointed out in the paper, substantially uniform illumination is possible with any 
of the three characteristic types of systems. Achievement of such uniformity is 
more difficult with the "motion picture" type. In most other respects, the two 
systems are about equally effective. 

MR. KELLOGG: I should like to bring out one point in regard to the history of 
the curved filament lamp. The number two type, or stereopticon type, as Mr. 
Carlson has pointed out, has proved to permit very large tolerances in the posi- 
tioning of the lamp as well as in the shape of the filament or source. As shown by 
Mr. Carlson's curves, even turning the filament axis from horizontal to vertical 
had very little effect on the distribution of light. In fact, those who have been 
working with that system had come to rely on these tolerances so far that we gave 
very little consideration to the lamp itself in trying to get uniformity of light. 
Our first efforts were concerned more with improving the lens system. 

One of the men with whom I have been associated, Mr. L. T. Sachtleben, began 
several years ago to make some studies closely similar to those Mr. Carlson has 
been describing. He found that although the position of the lamp and the shape 
of the source did not make much difference, the system is sensitive to changes in 


the intensity with the direction from which the lamp is viewed. He found that 
the lamps would measure brighter and dimmer as the direction of viewing was 
changed. This was because the inside of the filament helix looks brighter than 
the outside. In some positions the inside is hidden by the outside, and hi other 
positions both are in view. Mr. Sachtleben found that this was quite pronounced; 
the irregularities were made much worse by the fact that the hiding of the inside 
turns occurred almost simultaneously, due to the uniformity of spacing of the 
turns. He suggested that by curving the filament we could break up that uni- 
formity of spacing and get greater uniformity of illumination. That was tried 
out and it proved to be so satisfactory that it has become standard with us. 

I think Mr. Carlson would have mentioned this point himself except for the fact 
that it originated before his immediate contact with that phase of the problem, 
and, therefore, he would have had to state it on hearsay. 

MR. CARLSON: The curved filament construction is of definite value in im- 
proving uniformity of illumination with both the recording type of optical system 
to which Mr. Kellogg refers and the "stereopticon" type of reproducing system. 
It is the only construction in use which provides the desired uniformity of illumi- 
nation from the recording system in question. For reproducing systems of the 
"stereopticon" type, comparable uniformity is obtainable with the straight coiled 
filament and it is recommended because of its inherently lower cost and perform- 
ance characteristics. 


Summary. An explanation is given of lighting problems from the viewpoint of 
the cinematographer. Certain advances in equipment and working tools remain in 
obscurity for a long period before they find their rightful places in motion picture set 
lighting because they seem to interfere with dramatic effect. If they possess merit, 
however, they are gradually adapted to general use. A typical example is the light- 
meter, which is now going through the final stages of assimilation to studio lighting 
technic. New fast films have been brought into use and the resulting changes in 
lighting technic are now in the process of perfection. Recent changes in lighting 
equipment are described. Three new higher speed negative films for the Technicolor 
process are being used. The effect of the new films on Technicolor set lighting is 

The mechanics of motion picture studio lighting consist largely 
in placing illuminants around and above a set in such a manner as to 
obtain an overall light of sufficient intensity for desired exposure; 
to increase intensity on points of interest in order to relieve flatness 
and make interesting areas stand out; and with the aid of diffusers 
and black screens to diffuse and block out undesired light. 

With requirements that may be stated in one paragraph it would 
seem that knowledge of the latitude and speed of the film coupled 
with the use of a suitable light-meter should make studio set lighting 
a comparatively simple procedure. However, the mechanics of studio 
lighting, important as they are, sink into insignificance when com- 
pared to the main problem, which consists in using light as an aid to 
the dramatic effect of the photoplay. 

The dramatic effect in motion picture photography is a mysterious, 
intangible thing. Imagine the heroine of a photoplay on a set built 
to represent the interior of a haunted house. A circular staircase 
winds upward toward a cobweb-covered door. The lighting is 
arranged to create a chilly, ghostly atmosphere, yet the heroine must 
roam at will through the set without moving into an area where the 
light will make her appear other than attractive. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
17, 1939. 



The successful director of cinematography is a combination of 
engineer, artist, writer, director, and salesman. With his knowledge 
of story value and his ability as a psychologist he is telling his story 
with light just as the writer is with words. 

The director of cinematography must at once satisfy the director, 
producer, and often the star or stars of the picture. His prime 
interests are camera angles, mood, sweep, and many other problems 
serving to make the total dramatic effect. New types of motion 
picture film, cameras, lenses, lighting equipment, and light-meters 
are the tools that may make it possible to enhance this dramatic 
effect, but if they seem to interfere with the technic that has been 
developed through years of experience they sometimes go through a 
long period of testing before they are accepted. 

The laboratory technician may desire a negative showing some 
detail in all shadow and highlight areas with most of the density 
range on the straight-line portion of the gamma curve. The producer 
and art director may desire sufficient overall density to insure an 
appreciation of set values. Yet the cinematographer, director, or 
both, may desire a dramatic effect that fails to meet the requirements 
of either laboratory technician or art director, but does achieve the 
end result of affecting the senses of the theater patron in a manner 
which forwards the total desired effect of the story. 

If the cinematographer is able to expose the film in a technically 
correct manner, to show the cost value of the set, and still achieve the 
desired dramatic effect, he has accomplished the ideal. 

Mathematically two and two are always four; but the enemy of 
successful drama is strict formula. Thousands upon thousands of 
dollars are spent in an effort to twist the dramatic formulas to make 
them appear new and different to the viewing public. It is for these 
reasons that certain technical improvements sometimes find slow 

At this writing the studio lighting situation is in a somewhat 
unsettled condition. New and faster films have brought about a 
cycle wherein a great deal of interest is shown in learning just how low 
the set levels may be held, or as to just what photographic improve- 
ment may be obtained by maintaining the higher levels and greatly 
improving definition by reduced lens apertures. 

A continued trend toward the increased use of spotlighting equip- 
ment and away from floodlighting units has resulted in a technic 
in which points of interest are maintained at a level suiting the re- 


quirements of the cinematographer, and shadow areas are allowed to 
go dark. This has resulted in an increased use of small spotlighting 

In summing up the experiences of a number of cinematographers 
with the new films, it may be said that as many units are used as 
before to obtain the same general type of photography, but the in- 
tensity of the units has been decreased by the use of bulbs of lower 
wattage, smaller carbon arc units, by moving the lamps farther back, 
or by adding diffuser mediums. 

The exact amount of light reduction depends upon whether the 
cinematographer wants a sharp, crisp negative with translucent 
shadow areas, or a soft, "effect" type of lighting. Considerable light 
reduction has been accomplished in some cases, but the present 
trend seems to be toward higher levels. 

Past changes in film speeds have shown that a new technic of 
set lighting is developed only after the cinematographers have had 
sufficient time to test the practical application of the new emulsions 
by gradual exploration. Some portion of the additional film speed 
will probably be used to maintain reduced light levels, but continued 
experimenting will be carried on to determine how the additional 
speed may be properly used to improve photographic quality by the 
use of reduced lens apertures for increased definition. A proper 
light balance at a reasonably high level will simplify the problem 
of obtaining adequate screen light to satisfy the viewing public. 

Arc Lamps on Black-and- White Sets. The general use of arc 
lamps mixed with incandescents on black-and-white sets is well 
established. The proportion of arc lamps to incandescent units 
depends upon the desires of the cinematographer and the size of the 
set. In cases where the general level from incandescents has been 
reduced by using bulbs of lesser wattage, a light reduction from the 
arcs has been accomplished by using smaller units, moving the lamps 
farther back or by adding diffusers. 

Fluorescent Tube Lighting. White fluorescent tubes are being used 
in some close-ups to give soft front lighting and to "iron out" wrinkles. 
The new emulsions have made the use of this type of illuminant 

Light- Meters. The increased use of the photoelectric exposure 
meter on interior sets since the advent of the new films has been 
phenomenal. A paper published in December of 1938 forecast the 
possible use of a direct-reading, rather than a reflection-reading 


meter for use under artificial illumination. Several companies have 
produced such meters and they are being used generally. 

There is still some reluctance to the acceptance of light-meters, 
on the grounds that their use will restrict the cinematographer from 
using individuality in his work. This has not been found to be the 
case where light-meters have been used extensively. 

The most successful use of light-meters seems to be where the 
meter reading is taken from the plane in which the principal actor's 
face will be, with the meter pointed directly at the source of key 
light. After the key light has been established at the proper level the 
rest of the illumination is balanced visually. Sometimes the high- 
light and shadow areas are checked as an added precaution. The 
light-meter is particularly valuable when exploring the possibilities of 
a new film emulsion, or new types or arrangements of lighting units. 

Lighting for Technicolor. Technicolor has also introduced faster 
films, and what applies to black-and-white photography will likewise 
probably apply to the technicolor process. As this report is written 
there has not been sufficient footage of the new technicolor film ex- 
posed to establish a technic. 

C. W. HANDLEY, Chairman 




Summary. A report of the work of the Committee since the last Convention. 
Work on the proposed revision of the NFPA Regulations for Handling Nitrocellulose 
Motion Picture Film has been completed and the revision will be placed before the 
NFPA at the Chicago meeting in May. The present report discusses also the 
Committee's search for practicable and inexpensive light-measuring instruments for 
use in theaters, in addition to other subjects engaging the attention of the several 

A number of important matters are actually before the Com- 
mittee at this time. As detailed information on these matters is 
not yet complete, they will be reported on more fully at a later date. 
Among these projects are : 

(1) Study of screen brightness and methods of measuring it. 

(2) Problem of obtaining simple and practicable meters for measuring screen 
illumination and brightness. 

(3) Study of screen sizes and placement, and seating arrangement. 

(4) Study of tolerances and tensions permissible in motion picture projection 
equipment and means of measuring and checking the values. 

(5) Revision of the NFPA Regulations for Handling Nitrocellulose Motion 
Picture Film. 

(6) Survey of the power requirements of motion picture theaters. 


The Committee has had under consideration for quite some time 
the subject of measuring reflected light and while it is relatively easy 
to obtain meters which are calibrated correctly on diffused light, the 
readings obtained from motion picture screens are greatly influenced 
by the fact that all commercial screens are in some degree specular, 
that is, not completely diffusing. This fact precludes the possibility 
of easily making absolute brightness measurements and for this 
reason the Committee is having tests made on various types of screens 
to determine whether an empirical method of testing can be estab- 
lished and whether thereafter it can be systematically correlated to 
a primary and precision method yet to be devised. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif.; received 
April 5, 1939. 




The report of the Committee, published in the November, 1938, 
JOURNAL, contained the complete recommendations of the Com- 
mittee with regard to the revision of the Regulations. These pro- 
posals were submitted to the National Fire Protection Association 
and are now being considered by the NFPA Committee on Hazardous 
Chemicals and Explosives. It is expected that final action on this 
revision will be taken at the May convention of the NFPA at Chicago. 
The Committee feels gratified that a great number of its proposals 
have been accepted by the NFPA Committee without modification. 
As it has been many years since the last issue of the "Regulations," 
this revision fulfills an important need of the industry in bringing 
theaters up-to-date as regards equipment and installation. 


A great deal has been written in various non-technical trade 
publications on the subject of theater lighting and power equipment 
operating characteristics. Also many attempts have been made to 
relate the lighting and power equipment installations to total energy 
costs. Developments in the motion picture field have reached the 
stage where every important operation is related in some manner to 
electrical apparatus of widely varying types. Some exhibitors and 
projectionists do not have reliable sources of information to deter- 
mine whether or not their equipment and their methods of operation 
conform to present-day trends. There has long been a need for a 
comprehensive report showing the various types of electrical equip- 
ment, their load characteristics, and their use and cost of operation. 
The Committee now has in progress the preparation of such a re- 

HARRY RUBIN, Chairman 









Summary. A brief account of the work of the Committee during the past year, 
including handling of shipping cases, direction of rewinding film returned from 
theaters, disposition of scrap film, use of lacquer in splicing, etc. 

During the past year meetings of the Committee were held regu- 
larly each month at the Firenze Restaurant in New York, and 
although there is little material of a specific nature to be reported at 
this time, nevertheless these meetings have proved of great value in 
providing periodic contacts among the heads of the exchanges of the 
various companies, and to permit interchange of ideas and discus- 
sions of technic, administration, and conduct, to the general better- 
ment of exchange operation. 

Some of the studies initiated last year have not yet been completed. 
For example, the Committee is awaiting further reports from its 
members and from other Committees of the Society and the Academy 
of Motion Picture Arts & Sciences on the relative merits of various 
modes of processing release prints, and as soon as all the material is 
available the Committee will be able to report further on the subject. 

The problem of mutilation and mishandling of release prints has 
received very close attention by the Committee. Although the 2000- 
ft reel has received complete acceptance by the industry and has 
proved very successful, a number of minor problems have arisen dur- 
ing the period of adjustment of the industry to the new size. The 
weight of cases, for example, has met objection in some quarters and 
seems to have aroused a feeling that rough handling of the cases must 
be expected and tolerated as something inherent in the business. 
The Committee is trying to discourage this idea, and to show that 
what is now regarded as "ordinary" wear and tear should no longer 
be regarded as ordinary. Contacts have been made with the various 
carriers of film and good cooperation has been shown. It is expected, 
therefore, that some improvement may soon be evidenced in this 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 21, 1939. 



Recently the question has been revived as to which is the better 
way of shipping film from the exchanges to the theaters "heads up" 
or "tails up." About two years ago, the Committee made an exten- 
sive study of the subject, principally in the interest of determining 
the best method of splicing. Drawings and data obtained from the 
film stock manufacturers showed that the most satisfactory splice 
was made when the film was wound in the "head-to-tail " direction, 
that is, leaving the tail out. 

A survey indicated that almost 88 per cent of the films returned 
to the exchanges had the tails out. This meant one of two things, 
viz., either the majority of the theaters of the country were not 
equipped with special reels, or they could (or would) not spare the 
additional tune required for rewinding in order to return the film to the 
exchanges "heads up." 

The representatives of the various companies on the Committee 
felt that some expense might be involved if it were decided to adopt 
the "tails up" system, which involved an additional rewinding 
at the exchange. Nevertheless, they were willing to go to this ex- 
pense if, as a result, the quality of the films delivered to the theaters 
and the resulting projection should prove much better. However, 
in view of the overwhelming figures indicating that projection in 
general was definitely on a "tails-out-return-to-branch" basis, the 
project was abandoned, and since that time it has not been brought 
before the Committee again. 

With regard to the use of lacquer in splicing film, investigation 
showed that there was no uniformity of practice among the exchanges, 
some of them not using lacquer at all and others for first-run films 
only. The efficacy and necessity of using the lacquer are now being 
studied further by the Committee. 

The problem of disposing of scrap film was also under considera- 
tion, and a canvass of the companies was made to determine what the 
current practices were. It was found that general procedure in the 
exchanges was satisfactory and according to regulations, although 
differing somewhat in details. This brought up the question also 
of fire regulations in exchanges, and as it was reported that the 
National Fire Protection Association was engaged in revising its 
"Regulations for Handling Nitrocellulose Motion Picture Film," 
a Sub-Committee appointed for the purpose drew up a set of sugges- 
tions, relating to exchanges, for the consideration of the NFPA. 

Other subjects under consideration by the Committee include 


methods of blooping, dryness and brittleness of film, cleaning films 
in exchanges (several of the companies are now cleaning their films, 
with gratifying results), and handling and storing film cement. 
In addition, the Committee has investigated a number of new de- 
vices, such as film cleaners, metal and fiber reel bands, new designs of 
shipping cases and reels, etc. 

A. W. SCHWALBERG, Chairman 






Summary. A brief announcement of the present membership of the Society, and 
the growth during the past year. 

During the past year, the increase of membership has been some- 
what slower than the increase during the previous several years, 
although we are gratified to know that the membership is increasing, 
however, slowly. The total of the new members acquired is 59, 
and 12 old members were reinstated. Unfortunately, there were 
losses due to resignations and deaths totalling 48, which make a net 
total increase for the year of 23. At present the total membership 
consists of 6 Honorary members, 140 Fellows, 356 Active members, 
and 840 Associates, totalling 1342. Pending action by the Admis- 
sions Committee are 14 Actives and 11 Associates, upon admission 
of which the total will rise to 1367. 

The growth of membership during the past several years is very 
striking when compared with the membership in 1933. In 1931, 
the membership was 760, whereupon it dropped preciptantly to 560 
in 1933. However, from that time on, the growth has been very 
steady and almost linear, so that the present membership is two and 
one-half times what it was in 1933. 

Although during the several past years the subscriptions have been 
increasing steadily, 1938 showed a slight drop, the net decrease being 
61. The present total number of subscriptions is 328. 

E. R. GEIB, Chairman 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 7, 1939. 

























































5 1 

i i 







^ j 

















paper i 

e on a 1 






























The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, N. Y. Micro copies of articles in maga- 
zines that are available may be obtained from the Bibliofilm Service, Department of 
Agriculture, Washington, D. C. 

Acoustical Society of America, Journal 

10 (Apr., 1939), No. 4 
Mechanism of Hearing by Electrical Stimulation (pp. 261- 

269 ' S. S. STEVENS AND 


On Sound Localization (pp. 270-274) H. WALLACH 

Multitone (pp. 275-279) W. L. BARROW 

Absorption of Sound by Vibrating Plates Backed with an 

Air Space (pp. 280-287) R. ROGERS 

Theoretical Determination of Sound Absorptivities by the 

Impedance Method with Experimental Verification (pp. 

288-292) H. A. LEBDY 

Comparison of Sound Absorption Coefficients Obtained by 

Different Methods (pp. 293-299) F. J. WILLIG 

Reverberation-Time Scale for High Speed Level Recorders 

(pp. 300-301) K. C. MORRICAL 

Reverberation-Time Meter (pp. 302-304) W. M. HALL 

Multiple Coil, Multiple Cone Loudspeakers (pp. 305-312) H. F. OLSON 
Velocity of Sound in Air (pp. 313-317) W. H. PEILEMEIER 

Mechanical and Electrical Analogies of the Acoustical 

Path (pp. 318-323) H. L. SAXTON 

Reversed Speech (pp. 324-326) E. W. KELLOGG 

American Cinematographer 

20 (May, 1939), No. 5 
Hollywood Engineer Designs New Type Meter (pp. 200- 

201, 230) W. STULL 

To Use Blood in Color Photography (pp. 202-203) I. B. HOKB 

Making Stereoscopic 8-Mm. Pictures in Color (pp. 210-211) J. V. WOOD 
Fast Films and Color Have Made Big Light Changes (pp. 

213-214) G. KORNMANN 

Practical Gadgets Expedite Camera Work (pp. 215-218) G. TOLAND 
Shooting from Air Uncovers Camera Marvels (pp. 223-224) R. B. STITH 



British Journal of Photography 

86 (Mar. 31, 1939), No. 4117 
Progress in Colour (pp. 198-200) 

86 (Apr. 7, 1939), No. 4118 
Progress in Colour (pp. 211-213) 

86 (Apr. 14, 1939), No. 4119 
Facts and Assumptions Relating to Alkaline Development 

(pp. 230-231) Part 1 R. B. WILLCOCK 

86 (Apr. 21, 1939), No. 4120 
Progress in Colour (pp. 243-244) 

86 (Apr. 28, 1939), No. 4121 
Progress in Colour (pp. 259-260) 

British Kinematograph Society, Journal 

2 (Apr., 1939), No. 2 

Social and Political Aspects of Films (pp. 75-86) S. ROWSON 

Growth of Kinematograph Technique in Great Britain 

(pp. 87-98) L. H. BACON 

Some Acoustic Faults in Large Auditoria (pp. 98-106) H. J. O'DELL 
Sound Recording Developments in Germany (pp. 106-110) H. L. BOHM 
Standardization of Process Projectors (p. 110) 
Large-Screen Television Equipment for the Kinema (p. 

High Speed Kinematography in Post Office Engineering 

(pp. 119-124) R. W. PALMER 

Educational Screen 

17 (Dec., 1938), No. 12 

Motion Pictures Not for Theatres (pp. 325-328) A. E. KROWS 

18 (Jan., 1939), No. 1 
Motion Pictures Not for Theatres (pp. 13-16) A. E. KROWS 

18 (Feb., 1939), No. 2 
Motion Pictures Not for Theatres (pp. 49-52) A. E. KROWS 

18 (Mar., 1939), No. 3 

Motion Pictures Not for Theatres (pp. 85-88) A. E. KROWS 

18 (Apr., 1939), No. 4 
Motion Pictures Not for Theatres (pp. 121-124) A. E. KROWS 


12 (Mar., 1939), No. 3 
Cathode-Ray Amplifier Tubes (pp. 9-11, 76) 

International Photographer 

11 (Apr., 1939), No. 3 

Fundamental Photographic Physics (pp. 10-11) Part 2 D. HOOPER 
Projection Symposium (pp. 13, 16-17) Part 6 W. JONES 

International Projectionist 

14 (Apr., 1939), No. 4 
Supplementary Sources of Service Data Anent Sound 

Systems (pp. 7-10) A. NADELL 


The Epoch of Progress in Film Fire Prevention (pp. 10, 

21-26) A. F. SULZER 

Giant Twin Projectors Feature Novel Eastman Fair Ex- 
hibit (pp. 12-13) 


21 (Mar., 1939), No. 3 

Der vollkommene plastische Film (The Perfect Stereo- 
scopic Film) (pp. 61-67) W. HESSE 

Neuzeitliche Verstarkereinrichtungen fur das Tonfilm- 
Forschungslaboratorium (Recent Amplifying Arrange- 
ments for the Sound Film Experimental Laboratory) A. NARATH AND 
Part 2 (pp. 67-72) K. H. R. WEBER 

Welche Wege gibt es, um die Wiedergabe in Filmtheatern 
zu verbessern? (What is the Best Method of Reproduc- 
tion in the Theater?) (pp. 72-74) 

Kinotechnik auf der Leipziger Fruhjahrsmesse 1939 (Mo- 
tion Picture Apparatus at the Leipzig Spring Fair, 1939) H. FICHTNER 

Motion Picture Herald (Better Theatres Section) 

135 (Apr. 29, 1939), No. 4 

The Advantages of the Smaller Image for Black-and- 
White Pictures (pp. 39-40) F. H. RICHARDSON 


12 (Apr., 1939), No. 134 

French Progress in Television (pp. 196-198, 199) R. BARTHELEMY 

Television Picture Faults and Their Remedies (pp. 212- 

214, 220) Part V S. WEST 

Phillips Technical Review 

4 (Jan., 1939), No. 1 

A Film Projection Installation with Water-Cooled Mer- 
cury Lamps (pp. 2-8) G. HELLER 

Photographische Industrie 

37 (Mar. 1, 1939), No. 9 

Beitrag zur Entwicklung des Durchsichtssuchers (Work 
on the Development of the Eye Level Finder) (pp. 280- 
282, 284, 286-287) K. MARTINI 

RCA Review 

3 (Apr., 1939), No. 4 
Gamma and Range in Television (pp. 409-417) I. G. MALOFF 



Officers and Committees in Charge 

E. A. WILLIFORD, President 

S. K. WOLF, Past-President 

W. C. KUNZMANN, Convention Vice-President 

J. I. CRABTREE, Editorial Vice-President 

D. E. HYNDMAN, Chairman, Atlantic Coast Section 
J. HABER, Chairman, Publicity Committee 

S. HARRIS, Chairman, Papers Committee 

H. GRIFFIN, Chairman, Convention Projection 

E. R. GEIB, Chairman, Membership Committee 

Reception and Local Arrangements 

D. E. HYNDMAN, Chairman 







Registration and Information 

W. C. KUNZMANN, Chairman 


Hotel and Transportation 

J. FRANK, JR., Chairman 

C. Ross P. D. RIES 



J. HABER, Chairman 



P. A. McGuiRE 

P. A. McGuiRE 


Convention Projection 

H. GRIFFIN, Chairman 






Officers and members of Projectionists Local 306, IATSE 

Banquet and Dance 

A. N. GOLDSMITH, Chairman 





Ladies 1 Reception Committee 

MRS. O. F. NEU, Hostess 






Headquarters. The headquarters of the Convention will be the Hotel Pennsyl- 
vania, where excellent accommodations have been assured, and a reception suite 
will be provided for the Ladies' Committee. 

Reservations. Early in September room reservation cards will be mailed to 
members of the Society. These cards should be returned as promptly as possible 
in order to be assured of satisfactory accommodations. The great influx of visi- 
tors to New York, because of the New York World's Fair, makes it necessary to 
act promptly. 

Hotel rates. Special per diem rates have been guaranteed by the Hotel Penn- 
sylvania to SMPE delegates and their guests. These rates, European plan, will 
be as follows: 

Room for one person $ 3.50 to $ 8.00 

Room for two persons, double bed $ 5.00 to $ 8.00 

Room for two persons, twin beds $ 6.00 to $10.00 

Parlor suites: living room, bedroom, $12.00, $14.00, and 
and bath for one or two persons $15.00 

Parking. Parking accommodations will be available to those who motor to 
the Convention at the Hotel Fire Proof Garage, at the rate of $1.25 for 24 
hours, and $1.00 for 12 hours, including pick-up and delivery at the door of the 

114 1939 FALL CONVENTION [j. s. M. P. E. 

Registration. The registration desk will be located on the 18th floor of the 
Hotel at the entrance of the Salle Moderne, where the technical sessions will be 
held. Express elevators from the roof will be reserved for the Convention. All 
members and guests attending the Convention are expected to register and receive 
their badges and identification cards required for admission to all the sessions of 
the Convention, as well as to several de luxe motion picture theaters in the vicinity 
of the Hotel. 

Technical Sessions 

The technical sessions of the Convention will be held in the Salle Moderne of 
the Hotel Pennsylvania. The Papers Committee plans to have a very attrac- 
tive program of papers and presentations, the details of which will be published 
in a later issue of the JOURNAL. 

Luncheon and Banquet 

The usual informal get-together luncheon will be held in the Roof Garden of the 
Hotel on Monday, October 16th. 

On Wednesday evening, October 18th, will be held the Semi-Annual Banquet 
and Dance, also in the Roof Garden of the Hotel. At the banquet the annual 
presentation of the SMPE Progress Medal and the Journal Award will be made, 
and the officers-elect for 1940 will be introduced. 


Motion Pictures. At the time of registering, passes will be issued to the dele- 
gates of the Convention admitting them to several de luxe motion picture theaters 
in the vicinity of the Hotel. The names of the theaters will be announced later. 

Golf. Golfing privileges at country clubs in the New York area may be ar- 
ranged at the Convention headquarters. In the Lobby of the Hotel Pennsylvania 
will be a General Information Desk where information may be obtained regard- 
ing transportation to various points of interest. 

Miscellaneous. Many entertainment attractions are available in New York to 
the out-of-town visitor, information concerning which may be obtained at the 
General Information Desk in the Lobby of the Hotel. Other details of the enter- 
tainment program of the Convention will be announced in a later issue of the 

Ladies' Program 

A specially attractive program for the ladies attending the Convention is being 
arranged by Mrs. O. F. NEU, Hostess, and the Ladies' Committee. A suite will 
be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. Further details will be published in a succeeding 
issue of the JOURNAL. 

New York World's Fair 

Members are urged to take advantage of the opportunity of combining the 
Society's Convention and the New York World's Fair on a single trip. Informa- 
tion on special round-trip railroad rates may be obtained at local railroad tickt 

July, 1939] 1939 FALL CONVENTION 115 

offices. Trains directly to the Fair may be taken from the Pennsylvania Station, 
opposite the Hotel: time, 10 minutes; fare, 10. Among the exhibits at the 
Fair are a great many technical features of interest to motion picture engineers. 

Points of Interest 

Headquarters and branch offices of practically all the important firms engaged 
in producing, processing, and exhibiting motion pictures and in manufacturing 
equipment therefor, are located in metropolitan New York. Although no special 
trips or tours have been arranged to any of these plants, the Convention provides 
opportunity for delegates to visit those establishments to which they have entree. 
Among the points of interest to the general sightseer in New York may be listed 
the following: 

Metropolitan Museum of Art. Fifth Ave. at 82nd St.; open 10 A.M. to 5 P.M. 
One of the finest museums in the world, embracing practically all the arts. 

New York Museum of Science and Industry. RCA Building, Rockefeller Cen- 
ter; 10 A.M. to 5 P.M. Exhibits illustrate the development of basic industries, 
arranged in divisions under the headings food, industries, clothing, transportation, 
communications, etc. 

Hayden Planetarium. Central Park West at 77th St. Performances at 11 A.M., 
2 P.M., 3 P.M., 4 P.M., 8 P.M., and 9 P.M. Each presentation lasts about 45 minutes 
and is accompanied by a lecture on astronomy. 

Rockefeller Center. 49th to 51st Sts., between 5th and 6th Aves. A group of 
buildings including Radio City Music Hall, the Center Theater, the RCA Build- 
ing, and the headquarters of the National Broadcasting Company, in addition to 
other interesting general and architectural features. 

Empire State Building. The tallest building in the world, 102 stories or 1250 
feet high. Fifth Ave. at 34th St. A Visit to the tower at the top of the building 
affords a magnificent view of the entire metropolitan area. 

Greenwich Village. New York's Bohemia; a study in contrasts. Here are 
located artists and artisans, some of the finest homes and apartments, and some 
of the poorest tenements. 

Foreign Districts. Certain sections of the city are inhabited by large groups of 
foreign-born peoples. There is the Spanish section, north of Central Park; the 
Italian district near Greenwich Village; Harlem, practically a city in itself, num- 
bering 300,000 negroes; Chinatown, in downtown Manhattan; the Ghetto, the 
Jewish district; and several other such sections. 

Miscellaneous. Many other points of interest might be cited, but space permits 
only mentioning their names. Directions for visiting these places may be ob- 
tained at the Convention registration desk: Pennsylvania Station, Madison 
Square, Union Square, City Hall, Aquarium and Bowling Green, Battery Park, 
Washington Square, Riverside Drive, Park Avenue, Fifth Avenue shopping dis- 
trict, Grand Central Station, Bronx Zoo, St. Patrick's Cathedral, St. Paul's 
Chapel, Cathedral of St. John the Divine, Trinity Church, Little Church Around 
the Corner, Wall St. and the financial district, Museum of Natural History, 
Columbia University, New York University, George Washington Bridge, Brook- 
lyn Bridge, Triborough Bridge, Statue of Liberty, American Museum of Natural 
History, Central Park, Metropolitan Museum of Art, and Holland Tunnel. 


These films have been prepared under the supervision of the Projection 
Practice Committee of the Society of Motion Picture Engineers, and are 
designed to be used in theaters, review rooms, exchanges, laboratories, 
factories, and the like for testing the performance of projectors. 

Only complete reels, as described below, are available (no short sections 
or single frequencies). The prices given include shipping charges to all 
pouits within the United States; shipping charges to other countries are 

35-Mm. Visual Film 

Approximately 500 feet long, consisting of special targets with the aid 
of which travel-ghost, marginal and radial lens aberrations, definition, 
picture jump, and film weave may be detected and corrected. 

Price $37.50 each. 

16-Mm. Sound-Film 

Approximately 400 feet long, consisting of recordings of several speak- 
ing voices, piano, and orchestra; buzz-track; fixed frequencies for focus- 
ing sound optical system; fixed frequencies at constant level, for de- 
termining reproducer characteristics, frequency range, flutter, sound- 
track adjustment, 60- or 96-cycle modulation, etc. 

The recorded frequency range of the voice and music extends to 6000 
cps. ; the constant-amplitude frequencies are in 11 steps from 50 cps. to 
6000 cps. 

Price $25.00 each. 

16-Mm. Visual Film 

An optical reduction of the 35-mm. visual test-film, identical as to 
contents and approximately 225 feet long. 
Price $25.00 each. 







Volume XXXIII August, 1939 



Progress in the Motion Picture Industry Report of the Prog- 
ress Committee for the Year 1938 119 

Review of Foreign Film Markets during 1938. .N. D. GOLDEN 158 

Paramount Triple-Head Transparency Process Projector 

A. F. EDOUART 171 

Methods of Using and Coordinating Photoelectric Exposure- 
Meters at the 20th Century-Fox Studio D. B. CLARK 185 

A Sound-Track Projection Microscope G. M. BEST 198 

Further Improvements in Light- Weight Record Reproducers, 
and Theoretical Considerations Entering into Their Design . 

Current Literature 224 

Fall Convention 226 

Society Announcements 230 





Board of Editors 
J. I. CRABTREE, Chairman 




Subscription to non-members, $8.00 per annum ; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1939, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


** President: E. A. WILLIFORD, 30 East 42nd St., New York, N. Y. 
** Past-President: S. K. WOLF, RKO Building, New York, N. Y. 
** Executive Vice-President: N. LEVINSON, Burbank, Calif. 

* Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
** Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 

* Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 
** Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St... New York, N. Y. 

* Treasurer: L. W. DAVEE, 153 Westervelt Ave., Tenafly, N. J. 

** M. C. BATSEL, Front and Market Sts., Camden, N. J. 

* R. E. FARNHAM, Nela Park, Cleveland, Ohio. 

* H. GRIFFIN, 90 Gold St., New York, N. Y. 

* D. E. HYNDMAN, 350 Madison Ave., New York, N. Y. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* S. A. LUKES, 6427 Sheridan Rd., Chicago, 111. 

** H. G. TASKER, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1939. 
** Term expires December 31, 1940. 

Summary. This report of the Progress Commitee covers the period June, 1938, 
to April, 1939. The advances in the cinematographic art are classified as follows: 
(I) Cinematography: (A) Professional, (B) Substandard; (II) Sound Recording; 
(HI) Sound Reproducing Equipment; (IV) Television at the Close of 1938; (V) 
Publications and New Books; (VI) Appendix A: The Motion Picture Industry in 

Motion picture history for 1938 has repeated itself in that the most 
notable advances recorded during the year have, as in the previous 
year, been in the production of new panchromatic emulsions for pro- 
fessional cinematography. Of special interest is the new background 
fine-grain negative which permits considerable improvement in cine- 
matographic quality of background projection scenes. Outside this 
development, there is nothing of outstanding importance in either 
professional or amateur cinematography for the year 1938. 

In the field of sound recording and reproduction, progress during 
the year has been confined to improvement and modernization of 
equipment and technics rather than the introduction of any novel 
schemes. Considerable effort has been expended in bringing about 
a more uniform quality of sound projection in the theaters throughout 
the country by providing various test-films and making them avail- 
able to theater operating personnel. 

The Progress Report contains for the first time a complete section 
on television, and it seems quite probable that this section will be- 
come of more importance in future reports of the Progress Committee 
than it is at the present time. 

The committee wishes to thank the following companies for sup- 
plying materials and photographs for the report: Agfa Ansco Corp.; 
Bell & Howell Co. ; Warner Bros. Studio; Electrical Research Prod- 
ucts, Inc.; Eastman Kodak Co.; Ampro Corp.; General Electric 
Co.; Bell Telephone Laboratories; General Radio Corp.; Metro- 
Goldwyn-Mayer Studio; RCA Victor Corp.; and Twentieth Cen- 
tury-Fox Studio. 



J. G. FRAYNE, Chairman 







(I) Cinematography 

(A ) Professional 

(1) Emulsions 

(2) Cameras and Accessories 
(2) Stage Illumination 

(4) Color 

(5) Film Processing 

(6) Miscellaneous 

(B) Substandard 

(1) Films 

(2) Cameras 

(3) 16-Mm Sound Cameras 

(4) Projectors 

(5) Miscellaneous 

(II) Sound Recording 

(1) General 

(2) Equipment 

(3) Accessories 

(4) Recording Methods 

(HI) Sound Reproducing Equipment 

(IV) Television at the Close of 1938 

(1) Studio Pick-Up Equipment 

(2) Mobile Pick-Up Equipment 

(3) Transmitters 

(4) Signal Propagation 

(5) Receivers 

(6) Large Screen Pictures 

(V) Publications and New Books 

(VI) Appendix A 

The Motion Picture Industry in Japan 1938 


(A) Professional 

History of motion picture progress for 1938 has repeated itself, for 
again the outstanding advance has been in the improvement of pan- 
chromatic negative emulsions. 


(7) Emulsions. The negative photographic motion picture emul- 
sion has undergone a tremendous change since the introduction of 
the first panchromatic film in 1913. The major portion of this ad- 
vance in emulsion manufacture technic has occurred during the 
past seven years because these films did not vary much in their 
characteristics until 1931. In no year of the quarter century, how- 
ever, have so many new films appeared as in 1938. 

Another interesting trend of the times has been the ever-growing 
demand of the trade for more technical information with regard to 
film emulsions. This trend has been especially noteworthy through- 
out the past decade since the event of sound motion pictures. Each 
manufacturer has supplied the trade with detailed information when 
a new film was announced. 

In conformity with this established custom, the Eastman Kodak 
Company prepared a special technical pamphlet giving full data on 
their three new products, Plus-X Panchromatic Negative, Back- 
ground-X Negative, and Super-XX Negative, when these films were 
announced in October, 1938. Plus-X was shown to be definitely finer- 
grained and double the speed of standard panchromatic negative 
emulsions in use the previous year. The use of smaller stops with re- 
sulting improvement in image definition and a reduction in set illumi- 
nation were recommended with this film, which was stated to be 
especially suitable for all types of interior photography. 

Background-X was described as a faster emulsion than earlier films 
which were used for background work, and of equally as fine grain. 

Besides its satisfactory properties for use as a negative for projec- 
tion backgrounds, the film was claimed to be well suited as a panchro- 
matic negative exclusively for exterior photography. 

The third film, Super-XX, was stated to represent a film which had 
four times the speed of standard fast panchromatic emulsions in use 
the previous year. The increased speed was obtained without any 
appreciable graininess increase. This film was recommended for all 
types of photography under extremely poor lighting conditions. 1 

The Belgian firm Gevaert, Ltd., was stated to be planning to re- 
enter the professional 35-mm film field. A duplicating emulsion 
and a new positive film were said to be in production at the Antwerp, 
Belgium, plant of this company. 2 The Gevaert Panchromosa film 
was described technically in a French publication. It was said to 
possess greater red sensitivity and white-light speed than a previous 
product. 3 


A comprehensive account of the manufacture of motion picture 
film support and emulsions, coating of emulsions, and the operations 
of slitting, perforating, testing, and packing of finished film was given 
by Amor before the Royal Photographic Society and published in 
their Journal. 4 

A research program over the past 25 years to produce a suitable 
metallic film support was described by Carter. The resulting thin 
flexible non-ferrous metal with a chemically inert oxide surface was 
claimed to offer several advantages over cellulosic supports such as 
greater permanence, non-shrinkage, lighter weight, non-inflamma- 
bility, different images on two sides, etc. Projection by reflection with 
a loss of only 7 per cent of the light from absorption was claimed. 5 

(2) Cameras and Accessories. No outstanding improvement in 
professional camera design has been reported for 1938. The Mitchell 
NC type was somewhat improved and widely replaced many of the 
older types still in use in the studios. The one outstanding camera 
further improved but not yet on the market is the one built at 20th 
Century-Fox. That studio now has two in operation and others 
building, and is of such advanced design that its debut is eagerly 

The use of photoelectric exposure meters reached a new high during 
the year. They are gradually being adopted as a regular "tool" in 
most studios, as it has been proved then- scientific application results 
in a more uniform product, more easily handled by the laboratories, 
without taking away any of the individuality of the various camera- 
men. Progress may be predicted for the coming year, as an intensive 
study is being made, not only by the cinematographers, but by the 
Weston, General Electric, and other manufacturers of such meters, 
and the best methods to be used in measuring both incident and re- 
flected light. 

(3) Stage Illumination. The introduction of extremely fast emul- 
sions has resulted in the reduction of wattages used in electrical units, 
thus effecting a saving of almost one-half in current consumption. 
Also, the technic of lighting is undergoing a slight change, inasmuch 
as superfluous or "leak light" must be controlled, as it registers on the 
walls and subjects more quickly, due to the faster film. Spots are be- 
coming more prevalent, replacing gradually the more open lights, such 
as "broads," rifles, banks, etc.; the cinematographer has better control 
of his light with the more confined units. Mole-Richardson and 
Bardwell-McAllister have each brought out a new series of "baby" 


spots, Fresnel-lensed, more simply controlled and with a more uni- 
form field; they have also made improvements in their Junior and 
Keglite series, respectively. Bardwell-McAllister's Type T-5 Studio 
Spot, which substitutes a Fresnel lens for the spill-ring, should become 
very popular. 

Mole-Richardson's new, silent "Arc-Broad," developed for color 
work, is a splendid example of what scientific research, properly ap- 
plied, will do in solving the industry's peculiar problems. This lamp 
is silent; has a very even and intense field; is self-adjusting and will 
burn constantly for two hours, with practically no variation in in- 
tensity or color ; all requisites of lighting for color. 

(4) Color. In the field of two-color photography, the Eastman Ko- 
dak Company brought out a new pair of bi-pack negatives which 
they designate by the term "Zelcras." The new negative has con- 
siderably increased speed, but its biggest improvement lies in the 
semitransparency of the front emulsion, permitting much sharper 
back negatives with a corresponding increase in print definition. All 
two-color prints now being made are done on duplitized (double 
coated) stock which is made as a standard product by both the 
larger film companies. Cinecolor, Inc., and Consolidated Film In- 
dustries both produce two-color prints, and they both report increased 
volume during the year. The former reported that they did a total 
of about eight million feet, representing the capacity output of their 
old plant. They announce construction of a new plant in Burbank, 
Calif., to handle between fifty and one hundred million feet per year. 

Several different processes for two-color photography were an- 
nounced throughout the year. Of these Telco is unique in method 
and has been described in the press. 6 Another using a special camera 
attachment to produce two half-size images was tested on a short by 
RKO. Prints from these negatives were made optically and processed 
by one of the standard methods described above. 

In the field of three-color photography the increased speed of 
negative materials for black-and-white work was reflected in the 
Technicolor process. At the very close* of the year this organization 
brought out a new set of negatives which were expected to show from 
two to four times the speed of their standard stock. The negatives 
have not been used enough as yet to permit inclusion of further de- 
tails. All three-color prints made during the year were done by 
Technicolor. They report upward of one hundred million feet proc- 
essed. They also report completion of an English plant having a 



capacity of twenty-five million feet ahd additions to their Hollywood 
facilities to the extent of one hundred and thirty million feet capacity 
per year. 

The Dunning Process Company, Inc., announced a three-color 
camera employing three negatives in a special bipack combination 
and a processing method characterized by the fact that it is all 
photographic as distinguished from an imbibition process, such as 
Technicolor. They announced plans for increasing their facilities 
during the coming year. 





FIG. 1. 

Pantachrom subtractive lenticular bi- 
pack tricolor process. 

In October, Eggert of the Agfa Research Department, read a paper 
at the Berlin meeting of the Deutsche Gesellschaft fiir photograph- 
ische Forschung, on the Pantochrom subtractive lenticular bipack 
tricolor process. (Fig. 1.) The green and blue separation positive 
images are formed in a lenticular emulsion (nearest the lens) and the 
red separation image is formed in a single layer coating in contact 
with the emulsion side of the lenticular film. Positives are printed on 
double coated stock carrying on one side an ordinary silver bromide 
emulsion, and on the other a double coated emulsion having a purple 



pigment in the upper layer and a yellow pigment in the lower layer. 
Printing is effected by contact simultaneously from both camera 
films, the front one, which has gone through a reversal process, being 
printed on the double emulsioned side of the positive and the other 
camera film on the single emulsion side. The single emulsion of the 
projection print (carrying the sound-track) is processed by a catalytic 
bleach method to form a blue-green image, whereas the double emul- 
sion side is processed similarly to yield the remaining two-color sub- 
tractive positive. 7 

FIG. 2. Bell & Howell 35-mm non-slip sound printer. 

(5) Film Processing. In the professional field, the Bell & Howell 
)mpany has created a non-slip 35-mm sound printer which, though 
outcome of the principles recognized by RCA in 1936, offers many 
lovel features, every one and all of which assure fidelity of reproduc- 
tion and full protection of the negative sound record. The B & H non- 
slip printer operates in a horizontal position (Fig. 2), thus relieving 
both negative and positive film from unnecessary stress and keeping 
them free from possible damage which may be caused by lubricant 
materials, as it is impossible for the film to be contaminated by them. 
The design of the film-driving parts permits accommodation with- 


out slippage of two films whose length may differ from to 1.2 
per cent with respect to each other. The driving energy is supplied 
by a Ys-hp, 220- volt a-c, 3-phase, 60-cycle, 1725-rpm, ball-bearing, 
squirrel-cage induction motor, with heavy-duty worm-gear reducer 
to operate it at the recommended speed of 75 feet per minute, a 
mechanical filter insuring uniformity of motion. 

The printing aperture consists of an optical slit 0.005 inch in width 
produced by an optical system consisting of a prefocused 10- volt, 
7V2-ampere, vertical-filament exciter-lamp operating at approxi- 
mately 8 volts as a light-source, a short-focus condenser, an ultra- 
violet filter (with provision of rapid replacement so that any desired 
type of niters can be used) ; a mechanical slit 0.010 inch wide, and 
a fully corrected lens to focus the slit at the printing point at a re- 
duction of 2 to 1 in the form of a sharp optical slit 0.005 inch in width. 

FIG. 3. Sizes of fluorescent lamps commercially avail- 

(6) Miscellaneous. Byron Haskin, of Warner Bros. Studios, has 
developed and applied for patents on a triple-head background pro- 
jector. The method utilizes three projectors mounted on one center 
base, operating as a single unit and superimposing three identical 
pictures upon a single screen. 

This triple head projector provides greater illumination on present 
size screens, and permits the use of much larger screens in background 
projection, thereby greatly increasing the possibilities of process 
photography in color as well as in black and white. 

In the process background field, a twin-camera unit, properly co- 
ordinated, simultaneously makes a "plate" of double the ordinary 
width; when projected similarly with coordinated projection, the 
screen size is of double width, permitting a scope of action more in 
keeping with actual wide-angle non-process set-ups. 


The Fluorescent Lamp (Fig. 3). While the phenomenon of fluores- 
cence with ultraviolet light is not a new one, it was only during the 
past year that lamps involving this principle were made commercially 
available. They are all of tubular form, either one or one and a half 
inches, and available in 18, 24, 36, and 48-inch lengths. The exciting 
ultraviolet energy is obtained from a low-pressure mercury arc and 
by the use of phosphors of different composition, seven different 
colors including a good quality of daylight (color temperature, 
6500 K) are available. Daylight fluorescent lamps found immediate 
application in the motion picture studios for make-up room light- 
ing, in the mixing of paints, and in set painting, particularly where it 
was necessary to secure correct colors for color photography. A 
paper discussing the use of these lamps in studio applications is 
being given at this convention. 8 

(5) Substandard 

It is during only comparatively recent years that the tremendous 
possibilities offered by the motion picture have become fully realized. 
The realization of its possibilities and importance has created new 
and interesting uses for the motion picture, and has extended its 
scope into many new fields. This has stimulated the interest of the 
older, well established organizations and manufacturers to adapt 
their products to meet the changing requirements and to create new 
products to satisfy the new demands. It has also proved an incen- 
tive for new manufacturers to enter the field with a great variety of 
products and equipment, some of which have shown real merit. For 
the most part, however, these products could not be called new in 
the sense that the product is recently invented, novel, or progressive 
in contributing to the art of amateur cinematography. 

Mechanical equipment and adaptations in 1938 have far exceeded, 
in number, the introduction of chemical products intended for sub- 
standard cinematographic use, such as films. This is undoubtedly 
due to the fact that the manufacture of sensitized products in general 
requires knowledge, skill, and a capital investment that but few can, 
or would, care to bring together. 

Although many new films have appeared in 1938, most of them 
have been introduced by marketing organizations and the products 
represent the well known varieties of positive or negative film intro- 
duced to the public under new appellations. 

Abroad, as in immediately preceding years, 1938 has seen an in- 


crease in the popularity of American made equipment. Sixteen-mm 
projection equipment, especially, is in great demand. Sturdiness of 
mechanical design, high-powered illumination, together with ade- 
quate sound mechanisms, have made much of the better-grade 16-mm 
domestic equipment especially suited for hard, continuous use in 
theaters. For these reasons much of this type of equipment was 
exported to those countries where 16-mm films are regularly ex- 
hibited in theaters for entertainment purposes. 

(1) Films. Super XX, a new, high-speed, reversible 16-mm film, 
manufactured by the Eastman Kodak Company, was introduced 
during the latter part of 1938. 9 This new reversible film possesses 
four times the speed of regular panchromatic reversible material, and 
is about twice as fast as supersensitive panchromatic reversible 
films. This film greatly broadens picture-taking possibilities. 
Movies of stage performances, street scenes, and night indoor movies 
are now easily made. 

Agfa Supreme Negative, manufactured by the Agfa Ansco Corpora- 
tion, was made available in 16-mm width. This film fills the need of 
an extremely high-speed negative 16-mm material, and places in the 
hands of the 16-mm trade a film having properties identical to those 
of the films used by the professional 35-mm industry. Fine grain 
combined with brilliant gradation makes this new film a suitable 
recording medium for use wherever difficult lighting conditions are 
encountered, such as indoor sports, athletic events, and the like. 

Dupont Regular Panchromatic, a product of the Dupont Film 
Manufacturing Corporation, was announced. This film has a color- 
sensitivity closely approaching that of the human eye, and will give 
excellent results outdoors and is also suitable for use indoors with 
artificial illumination. Because of its complete color-sensitivity, a 
wide range of niters may be used. This film is of the reversible type, 
and is processed by the manufacturer using the Dupont reversal 

Agfa Hypan reversible film was made available in 50-foot cassettes 
for the Siemens-Halske camera. 

Service for duplicating 16-mm sound-film was announced by the 
Agfa Ansco Corporation through their New York City laboratory. 
Special equipment is used to provide sound duplicates of continuous 
length and uniform quality. 

An outstanding contribution to 16-mm color was made during the 
past year by an announcement by the Eastman Kodak Company of 



Kodachrome duplicates. 10 Amateurs, advertisers, and scientists have 
looked forward to the time when duplicates could be made from their 
Kodachrome films. Although this service was announced late in 
1938, it is already finding wide application. Technicolor is now 
prepared to make 16-mm Kodachrome copies of any of its 35-mm 
material. The Dunning Process Company makes copies of both 
35-mm two-and three-color originals and 16-mm Kodachrome origi- 
nals, both sound and silent; and the Stith-Noble Company also 
copies Kodachrome. The number of processing organizations in 
this field is increasing from time to time. 

FIG. 4. Filmo 141-B camera. 

(2) Cameras. Filmo 141, a 16-mm magazine camera, was an- 
nounced by Bell & Howell of Chicago. (Fig. 4.) This camera uses 
50-ft 16-mm film magazines. A new type of view-finder gives a mag- 
nified, sharply defined field. The front element of the view-finder 
unscrews for quicker interchange with others to match seven lens 
focal lengths from 15 mm to six inches. This camera is also equipped 
with a soft rubber cup over the eyepiece to prevent side glare, and 
makes the finder easier for wearers of glasses to use. The Filmo 
141 is available in either of two types, being identical except for 
speed range. The Filmo 141 -A has speeds of 8, 16, 24, and 32 frames 
per second, while the Filmo 141-B has speeds of 16, 32, 48, and 64 
frames per second. Either of these cameras may be operated at speeds 
intermediate between those calibrated speeds merely by setting the 


dial accordingly. In addition, the camera is also provided with a 
single-exposure device. 

A new Filmo 8-mm camera with turret lens mounting accommo- 
dating three lenses was introduced. (Fig. 5.) This new camera 
combines economy with features which heretofore have been avail- 
able only to users of 16-mm and 35-mm film. A novel feature of this 
camera is in the view-finder objectives, which are mounted on the 
turret. When a lens is in the photographing position, its matched 

FIG. 5. Filmo turret-8 camera: the "Aristocrat." 

finder objective is always in view-finder position. This is the first 
8-mm camera to be equipped with the positive type view-finder pro- 
viding a real image in the position of the view-finder mask rather than 
a virtual image in front of the camera. Since the image and the mask 
coincide with one another in position, the image never shifts. It 
always shows exactly as much of the subject as will appear on the 
screen, no matter at what angle the eye looks into the eyepiece. The 
image is full size without masking for telephoto lenses. The camera 
is equipped with a Taylor, Hobson l2 1 / 2 -mm, //2.4 lens. 


A new 8-mm camera manufactured by Ditmar abroad and offered 
in this country by Hans Unfried of Buffalo, N. Y., made its appear- 
ance in 1938. This new camera employs a built-in, photoelectric ex- 
posure meter. The scale of the exposure meter is viewed through 
the regular image view-finder. Beside the scale is an indicator show- 
ing the lens stop values. This is the first 8-mm camera making its 
appearance in the United States having a built-in exposure meter. 
The camera has two speeds, 16 frames per second and 32 frames a 
second. A change to either speed may be made instantly by pressing 
the proper button conveniently located on the camera. The camera 
is regularly equipped with an //2.8 lens. Higher-speed lenses and 
lenses of different focal lengths are also available. Models of this 
manufacturer were introduced last year, but did not have the built-in 
exposure meter. 

An 8-mm-camera, under the name of Eumig C-4, recently appeared 
on the American market. This camera is driven by a midget electric 
motor, power for which is supplied by flashlight batteries. The 
camera is box-like in appearance and weighs, complete with battery, 
about one and one-half pounds. The camera is equipped with an 
//2.5, 12V2-mm lens of the universal focus type. It operates at a 
film speed of 16 frames per second. 

Emel, an 8-mm cine camera, was introduced in France during the 
past year. This camera is one of the first 8-mm cameras to incor- 
porate professional features such as provision for lap dissolves and 
double exposures. The Model C-61 is equipped with a fixed focus 
//2.5 lens. The Model C-82 is equipped with a turret head for ac- 
commodating three lenses. The camera has provisions for exposing 
single frames. It has five speeds: 8, 16, 24, 48, and 64. 

A new Movikon 16-mm camera was introduced abroad by Zeiss- 
Ikon. The camera is of very compact design and is of the magazine 
type. It is equipped with a 20-mm, //2.7 Tessar lens. Provisions 
are made for interchanging lenses of various focal lengths. The 
camera has film speeds of from 8 to 64 frames per second. 

Siemens-Halske introduced a new 16-mm camera abroad having 
the range-finder and the view-finder combined in one unit. 

(3) 16-Mm Sound Cameras. To meet the demands of the indus- 
try for better 16-mm sound, Electrical Research Products, Inc., has 
designed a film-recording machine for recording directly on 16-mm 
film. The machine is intended primarily for studio use and its de- 
sign is such as to permit recording 16-mm variable-density negatives 



from direct pick-up or by re-recording from 35-mm film. Release 
prints are then obtained by contact printing the sound-track in com- 
bination with optical reduction printing of the picture from the 35-mm 

The sound-track is recorded by a recently developed variable- 
intensity light-valve modulator shown in the right-hand compart- 
ment of Fig. 6. The film is exposed at the periphery of a film-driven 
oil-damped kinetic scanner resulting in uniformity of film motion 
comparable to that of the latest 35-mm film-recording machines. 
The film-propelling mechanism is shown in the left-hand compartment 

FIG. 6. Variable-intensity light-valve modulator. 

of Fig. 6. A transparent deflector diverts part of the modulated 
light to the photoelectric cell and monitor amplifier located in a rear 

With proper equalization to compensate for the known losses, 
positive prints obtained from negatives recorded on this machine 
show a frequency characteristic substantially flat up to 6000 cycles. 

A new 16-mm sound-film camera was introduced by the Berndt- 
Maurer Corporation. The equipment is designed for single-system 
sound recording. The camera is motor-driven, and has a galva- 
nometer and an optical system which produce the "bilateral" or sym- 
metrical type of variable-area sound-track. The film motion of the 


camera is reported to be exceptionally steady, making the camera 
suitable for recording music as well as speech. Critical focusing is 
provided by a ground-glass viewer with an enlarging microscope. 
The camera is equipped with a built-in dissolving shutter mechanism. 
Controls for the camera are grouped on the back. Detachable maga- 
zines having either 400 or 1000-ft capacity are provided. 

FIG. 7. Kodascope Model G. 

Abroad, Zeiss Ikon have introduced the Ikophon, a 16-mm camera 
for single-system picture and sound recording. 

(4) Projectors. A new 16-mm projector was introduced by the 
Eastman Kodak Company. This projector is known as tfre Model 
G. (Fig. 7.) A special feature of this projector is the wide selection 
of lenses of various focal lengths which are available. It is readily 
adaptable to either 400, 500, or 750-watt lamps. This wide selection 
of lenses of various focal lengths and lamps of various intensities 



makes the equipment very flexible for meeting any projection condi- 

The Ampro Corporation introduced two new 16-mm sound pro- 
jectors (Fig. 8), Model X and Model Y. Both models are adaptable 
to either 750 or 1000-watt lamps. These new projectors are of 
compact design. Operation has been simplified to the extent that 
they are especially suitable for use by inexperienced operators, such 
as for classroom use, etc. The Model Y is suitable for either a-c or 

FIG. 8. Ampro Models X and Y. 

d-c operation, and will project at either silent or sound-film speed. 
Ampro also introduced the Ampro Arc. (Fig. 9.) This projector is 
especially suitable for displaying 16-mm motion picture films before 
large audiences, as in theaters. It is claimed that the high-intensity 
arc produces five times the screen brilliance of a 750-watt projector. 
This new projector has numerous special features, including a high- 
intensity type lamphouse, automatic carbon feeding, full-wave rec- 
tifier and collapsible type projection stand. The complete equip- 



ment includes the projector unit equipped with a 1600-ft reel ca- 
pacity and two torpedo speakers with tripod stands. 

Two new Filmo sound projectors were introduced by the Bell & 
Howell Company of Chicago. (Figs. 10 and 11.) The new models 
are known as the "Commercial," shown in Fig. 10, and the "Acad- 
emy," shown in Fig. 11, models. The "Commercial" model is a 
single-case machine, designed for use by industrial firms requiring a 
low-priced, sturdy, durable sound projector. The machine is readily 

FIG. 9. Ampro Arc. 

>rtable and has sufficient illumination and volume for effective 
mnd projection with audiences up to five hundred people. The 
"Academy" model is a two-case projection equipment, with the pro- 
ctor operating in a "blimp." The projector has a governor-con- 
)lled silent speed as well as a sound speed. In the "Academy" 
lodel are incorporated all the elements that contribute to satisfac- 
>ry projection on both sound and silent classroom films. Sim- 
)licity is the keynote of the "Academy" sound projector. 
The Filmoarc, a high-powered 16-mm projector, was introduced 



by Bell & Howell. This projector presents to the 16-mm industry a 
high-powered, sturdy projector suitable for hard continuous use in dis- 
playing pictures before large audiences heretofore unable to be served 
with 16-mm projection equipment. The new projector is equipped 
with an automatic carbon arc and mirror reflectors. 

Standard Projectors, Incorporated, introduced their Model 90-S. 
This silent projector is equipped with a 750-watt lamp and is said to 
have an unusually efficient optical system. This, together with the 
//1.6 projection lens, insures brilliant screen illumination. The pro- 

FIG. 10. Filmosound "Commercial" model. 

jector is equipped with a removable film-gate, framing device, and 
pilot light. The entire projector is of die-cast construction. All 
electric controls are centralized. The projector is equipped with 
forward and reverse motion and has a rapid mechanical rewind. 

American Bolex Company, Inc., introduced in the United States 
a new Bolex projector which has enjoyed much popularity abroad. 
A novel feature of this new machine is that it is interchangeable for 
projection of either 16-mm or 8-mm films. 

Pathe, Limited, introduced the Pathescope Model H, a 9V2-mm 



projector. This projector is adaptable for a-c circuits only, and is 
equipped with low- voltage and a high-amperage light-source operating 
through a transformer. The projector will accommodate reels up to 
300-ft size. 

Siemens-Halske introduced a new Siemens 8-mm projector. The 
projector has a 200-watt lamp and an adjustable film speed, varying 
from ten to twenty frames per second. The machine is equipped 
for still-frame projection. The projector is equipped with either a 
25-mm or 35-mm projection lens. 

FIG. 11. Filmosound "Academy" model. 

The firm of Heurtier, in France, introduced a new motion picture 
projector which will accommodate either 8-mm, 9V2-nim, or 16-mm 
film. The change from one gauge to another has been simplified to a 
great extent by the novel form of sprocket mounting used in the 
equipment. Both the feed and take-up sprockets are built on a ro- 
tating plate. A turn through 120 degrees brings one of the three 
(8-mm, 9V2-mm, or 16-mm) sprockets mounted onto the plate into 
position for carrying the film. The gate is easily interchangeable. 
In order to allow for the projection of old or very shrunken films, the 
pull-down mechanism is of the somewhat unusual triple-claw type. 
With 8-mm film, an extra short focal length lens is recommended; 


however, for 9V2-nim and 16-mm projection, a longer focal length 
lens is recommended, but the same lens can be used for either 9V2-mm 
or 16-mm films. 

(5) Miscellaneous. General Electric Company introduced a newly 
designed photoelectric exposure meter adaptable for all photographic 
use, especially recommended for cine work. The new meter is said 
to give an unusually accurate indication of exposure with very low 
light intensity. 

FIG. 12. Eastman reflex finder magnifier. 

A new Kodak 16-mm enlarger was introduced by the Eastman 
Kodak Company during the past year. This device makes it pos- 
sible to make black-and-white prints 2 x /2 by 3 3 /8 inches from either 
16-mm black-and-white films or from Kodachrome. 

The Eastman Kodak Company also introduced a new Kodascope 
movie viewer. The movie viewer shows movies on a built-in screen 



as the film is drawn through. This device assists in simplifying the 
editing of 16-mm motion picture film. 

A new Reflex Finder Magnifier for the Cine Kodak has been made 
available. (Fig. 12.) This finder shows on a ground-glass screen, 
working in conjunction with a built-in magnifying glass, the exact 
field of focus with whatever lens is being used. There is also avail- 

FIG. 13. 

Bell & Howell 8-mm titler, with Filmo 134 
"Sportster" camera. 

able the optical finder which shows the field of any lens and corrects 
for parallax down to two feet. 

A new 8-mm titler for Filmo cameras has been introduced by Bell 
& Howell. (Fig. 13.) This unit, comprising the main titler assembly 
and a cross-arm bracket bearing two sockets and reflectors, is equipped 
with a special copying lens, making it unnecessary to use the camera 
lens for a titling lens. The copying lens is corrected for either color 



or black-and-white film, giving remarkable sharpness and flatness of 
field very necessary in making 8-mm titles. 

An 8-mm editor has been made available by Bell & Howell. The 
new editor (Fig. 14) comprises a splicer and film-viewing device. 
The entire equipment is based on the design of the well known Bell 
& Howell laboratory splicing equipment. 

Projection Lamps (Fig. 15). A new 1000-watt, 110-120- volt, T-12 
bulb projection lamp intended primarily for high-power 16-mm mo- 
tion picture projection, was introduced by the two Mazda lamp 
manufacturers. This lamp incorporates several rather novel de- 
sign features, with the result that it delivers 50 per cent more light 
to the screen than do lamps of the old design and the same wattage. 

FIG. 14. Bell & Howell 8-mm editor. 

The Reflector Photoflood Lamp (Fig. 16). This very useful lamp 
for amateur motion picture photographers takes advantage of a re- 
cently developed process of depositing a layer of aluminum on the in- 
terior wall of a glass lamp bulb to form an efficient reflector. The 
base half of the bulb has been given a modified parabolic shape and 
aluminized to redirect the bulk of the light within an approximate 
angle of 60 degrees. The front half of the bulb is frosted inside 
to improve the uniformity of the illumination. The lamp is intended 
to be used in floor and table lamps and with a simple swiveling socket 
and clamp where separate reflectors are not feasible, for amateur 
still and movie photography, and as a supplementary light for pro- 
fessional photographers. It has a rating of 500 watts and six hours' 
life, the same as the number 2 Photoflood. 




(1) General. Existing sound recording systems have been care- 
fully examined during the past year for factors contributing to the 
impairment of sound quality and volume range. Special equip- 
ments 11 have been devised to investigate distortion of various forms 
occurring in recording and re-recording channels. The data obtained 
have exonorated certain parts of the system, but have suggested 
possible improvements, primarily in the link from the modulator to 
the reproducing machine PEC cell. Improvements in quality, and 
in overall volume range have been effected by push-pull modula- 
tors, 12 - 13 improved noise-reduction methods, pre- and post-equaliza- 
tion, track squeezing, 14 and volume compressors 15 and limiters. 16 

(2) Equipment. In the record- 
ing channel field, Electrical Re- 
search Products, Inc., in coopera- 
tion with Metro-Goldwyn-Mayer 
Studio, has developed a new 
simplified channel for production 
dialog. It consists of two units a 
combined amplifier - noise - reduc- 
tion and a film recorder equipped 
with all necessary motor controls. 
Both units are compact, and weigh 
40 and 100 pounds, respectively. 
Power is supplied from batteries or 

RCA have converted some of 
their variable-area recorders by 
changes in the optical system to 

record variable-density track. In addition, an optical volume limiter 
is being incorporated in some of their modulators. 

The use of various forms of volume compressors and volume limiters 
for dialog production is increasing. Advantages claimed are greater 
freedom from overload with corresponding improvement in quality 
and intelligibility. 

The use of new optical and modulator systems, and new type of 
film have brought about a demand for an increase in light for exposure. 
The high-pressure mercury lamp 17 developed by General Electric 
fulfills this need. 

With the increasing use of rectified a-c for power supply, several 

FIG. 15. 1000-watt, T12 bulb, 100- 
120-volt projection lamp. 



manufacturers have developed both low and high-voltage units in- 
corporating a self-regulating feature. These provide a constant 
output voltage for a wide range of current demand, and also com- 
pensate to some extent for a-c line voltage variation. 

An interesting method of introducing artificial reverberation into 
re-recording has been employed by M.-G.-M. An "echo pipe," a pipe 
some 300 feet long, is used. A moving-coil type of loud speaker fur- 
nishes acoustic energy to the pipe, and several microphone pick-up 
points are located at intervals along its length. Due to the finite 
time of transmission through the pipe, and the resulting reflections 
from its end, a condition simulating reverberation is produced. 
(3) Accessories. Two new microphones have been made avail- 
able. The first, by Bell Tele- 
phone Laboratories, is the 639- A 
cardioid directional microphone. 
(Fig. 17.) While intended pri- 
marily as a unidirectional device, 
each of its constituent elements 
may be used separately, thus 
providing bidirectional and essen- 
tially nondirectional response 
characteristics. The ribbon 
structure is of novel form and is 
relatively unsusceptible to me- 
chanical vibration. The second 
microphone, announced by 
ERPI, employs a miniature con- 
denser transmitter, vacuum tube, and transformer developed by 
Bell Telephone Laboratories. These components make possible a 
small unit of high quality. In size it is approximately seven inches 
long, two and a half inches in maximum diameter, and weighs one 
and a quarter pounds. 

Among new test equipment, General Radio Co. offers the 7 '36- A 
wave analyzer. (Fig. 18.) It is useful in determining harmonic dis- 
tortion in recording systems, and offers the following advantages over 
previous equipment: It is a-c operated, has a band-pass character- 
istic, and is stable and easily operated. The 760-A sound analyzer 
employs a novel method of securing a variable band-pass selection. 
It is obtained by placing a variable bridge circuit in the feedback 
path across a high-gain amplifier. In addition, the output meter 

FIG. 16. Reflector photoflood lamp; 
500-watt, 6 hours life. 



operates from a logarithmic circuit and can be read directly over a 
42-db volume range. Such an instrument, in conjunction with a 
sound-level meter, 18 is useful in evaluating the frequency spectrum 
of camera or projector noise. 

The modulated-carrier oscillator, which was developed during 
1937, has been reduced to a production design and a quantity have 
been manufactured. (Fig. 19.) This unit has become an excellent 
tool for the determination of 
optimum processing conditions 
for variable-area recording and 
for the maintenance of accurate 
control of film-processing labora- 

(4) Recording Methods. War- 
ner Bros. Studios have adopted 
push-pull variable-area recording 
for all original recording. Non- 
linear apertures were installed on 
all machines for original record- 
ings. These apertures are linear 
for 80 per cent modulation, but 
beyond this level require an 8-db 
increase to reach 100 per cent 
modulation, affording 6-db vol- 
ume compression, which is very 
desirable on shouting and ex- 
tremely loud dialog. The same 
studios are using fine-grain du- 
plicating stocks for re-recording 
prints and for the master prints 

FIG. 17. 

639-A cardioid directional 

from which the sound and picture 

negatives for foreign release are 

made. Great improvement in sound and picture quality has been 

realized as a result of the use of the new stock. 

Twentieth Century-Fox Studios have replaced the RCA variable- 
area system with a variable- density system of recording, wherein the 
usual galvanometer is used to move the penumbra of light with re- 
spect to an aperture, thus passing light onto the film with intensity 
varying with the deflection of the galvanometer. The usual monitor- 
ing card is used for determination of noise-reduction adjustments, etc. 



In order to improve the quality of Variable-area dialog recording, 
the RCA-equipped studios are employing an electronic limiter, tak- 
ing effect at a level of about 80 per cent modulation. In addition, a 
volume compressor is used both in original and dubbing recording. 
The compressor is likewise an electronic device, and may be adjusted 
to compress at different levels and different rates. The result is a 
reduction in objectionable overload for the same average level of 
speech. This device was initiated by J. O. Aalberg of RKO Studios, 

FIG. 18. General Radio type 736A sound analyzer. 

and has been universally accepted in all RCA variable-area installa- 

The use of squeeze-track in release prints has grown considerably 
during the year, being in use now at Universal, Columbia, and 
Paramount, in addition to M-G-M, the originator of the method. 

The intermodulation test has been introduced for control of 
variable-density recording, and has been used not only in studios, but 
in laboratories such as Paramount's. By means of this test, optimum 



negative and positive densities and gammas may be determined 
dynamically. The test consists of first recording a combination of 
low and high frequencies, such as 50 and 1000 cps, to be subsequently 
reproduced through a system which filters out the low-frequency com- 
ponent and measures the relative modulation of the remaining high- 
frequency at the low-frequency rate. 

A recording optical system has been developed that makes orignal 
push-pull recordings of the direct positive type. Anticipation of the 
recorded sound-waves by the noise-reduction system is made possible 
by providing a separation between the modulation and noise-reduc- 
tion light-beams. The optical system is being tested under com- 

FIG. 19. Modulated carrier oscillator. 

mercial conditions and has the advantages of less clipping, lower noise, 
and freedom from possible printer distortion. 


During 1938 considerable time was spent by the suppliers of re- 
producing equipment with members of the Academy of Motion Pic- 
ture Arts and Sciences participating in a cooperative program de- 
signed to improve the quality of sound in theaters. As a first step 
in its investigation, the Committee appointed by the Academy in- 
augurated a series of tests to determine upon a standard electrical 
characteristic for theaters which would present the recorded product 
of all the studios to the public to the best advantage and which, in 
addition, would fit the acoustic characteristic of a majority of the 


theaters. These tests were conducted employing the newer two- 
way type of loud speaker systems, typical of which is the RCA com- 
bined low and high-frequency units shown in Fig. 20. 

The characteristics ultimately decided upon have already been 
published. 19 Obviously, this characteristic is not a "cure-all," and 
may have to be modified depending upon the particular theater in- 

As a result of the tests conducted jointly with the Academy, the 
RCA Manufacturing Co. took the necessary steps to have the PG-90 

FIG. 20. Loud speaker combination for 

series of equipments so modified in the factory that they could be 
shipped with the desired characteristic. 

The above developments were a long step forward in the short but 
lively history of sound-on-film, but much still remained to be done in 
refining these developments and in bringing about other practical 
improvements, which, while probably not as revolutionary as a rotary 
stabilizer, were nevertheless important to the progress of sound 
projection in the theaters. 

The requirement for a correct two-way speaker system with a true 
cellular horn in the above field brought about the production by RCA 
of the PG-138 equipment. This equipment consists essentially of 


the MI-1040 sound-head (Fig. 21), the MI-9250 main amplifier, the 
MI-9512 field supply, the MI- 1444 low-frequency speaker, MI -9457 
low-frequency baffle, and the MI -1443 high-frequency speaker and 
MI -9485-6 high-frequency horn. 

All parts of the amplifier are conveniently arranged for ease in 
servicing. It is equipped with separate main system and monitor 
volume controls. The crossover network with standby switch is also 
contained in this cabinet. 

In emergencies the low-frequency horn can reproduce the higher 
frequencies, in addition to the low frequencies. A minimum of back- 
stage space is required for installation. The high-frequency horn is 
a true cellular design and provides equal distribution of the sound 
throughout the auditorium. It has the added and very distinct 
advantage that it can easily be adjusted for phasing and distribution. 

FIG. 21. MI-1040 sound head for PG-105-138. 

Realizing the necessity for a high-quality low-cost equipment for 
the medium size theater gave rise to the scheduling by RCA of the 
PG-139 equipment. This equipment includes a complete line of 
reproducing equipment, including sound-head, main amplifier, 
speaker field and lamp supply, pre-amplifier and low and high-fre- 
quency units. 

RCA have also introduced the PG-140, -141, and -142 series of 
equipments representing an outstanding achievement in sound repro- 
ducing equipment. They meet the specifications of the Academy of 
Motion Picture Arts and Sciences and have incorporated in their de- 
sign the refinements desired in earlier equipment of all makes. 
Thought was given in their planning to the requirements of the pro- 
jectionists, and nothing was omitted to give the exhibitor the best 
possible performance. 



A few of the features of these equipments are: Oil collection and 
mounting plate which makes the projector head easily adaptable to 
the sound-head and affords an effective oil seepage system. Isolated 
constant-speed sprocket drive reduces flutter. Micrometric optical 
system adjustment; double exciter lamp socket with prefocused 
lamps; optical system dowelled in position insuring positive azimuth 
adjustment. It will be observed in Fig. 22 that all units are in one 
rack, requiring minimum space and offering 
modern styling. 

The H-6 Water Cooled Mercury Lamp 
(Fig. 23). This lamp presents the interest- 
ing contrast of a large amount of light from 
a very small unit. The lamp itself is about 
three inches long and a quarter of an inch in 
diameter. The lamp is rated at 1000 watts 
and the light output 65,000 lumens. This 
comes from a source one inch long and Vis 
inch in diameter, giving a source brightness of 
over 193,000 candles per square-inch. The 
lamp must necessarily be cooled by a rapidly 
flowing stream of water. The quality of 
its light is such that it is not at present 
suitable for either motion picture photog- 
raphy or projection of colors. However, 
it is finding application as either a projec- 
tion or general lighting source where a 
powerful actinic light is required. The H-6 
lamp and its potential applications have 
been described in considerable detail in the 



The year was marked by refinement of all 

parts of the system in a steady advance toward commercial television. 
Late hi the year it was announced that a limited program service would 
be inaugurated in New York City with the opening of the 1939 World's 
Fair in that city, and that receivers would be offered for sale to the 
public at that time. Apparatus is available for sale to broadcasters 
for studio and transmitter service. Splendid progress was made through 
industry cooperation in establishing operating standards for a tele- 
vision system. Emphasis was placed on comprehensive field tests. 

FIG. 22. MI-9210 ampli- 
fier for PG140. 


(1) Studio Pick-Up Equipment. Steady progress has been made 
in the electrical and mechanical design of pick-up equipment for 
studio use. The frequency-band passed by the entire system has 
been widened and the circuit operation made more stable. The 
camera pre-amplifier and Iconoscope coupling circuits have been 
improved so that the signal-to-noise ratio has been increased. Operat- 
ing technic has constantly improved so that more consistent per- 
formance is obtained. Much attention was given to the problems 
of program production. 

(2) Mobile Pick-Up Equipment. Mobile pick-up equipment 

FIG. 23. Type H-6 water-cooled mercury vapor 
lamp and water jacket. 

mounted in trucks has been put in experimental operation and has 
given satisfactory performance for preliminary tests. The equip- 
ment includes an ultra-high-frequency transmitter for relaying the 
picture signal to the television transmitter for broadcasting. Loca- 
tion pick-ups have been successfully accomplished over moderate 
distances, and in one instance up to 27 miles. 

(3) Transmitters. Considerable improvement has been made in 
television transmitters. The modulating frequency characteristic 
has been widened. Circuits for inserting the direct current com- 
ponent in the transmitted signal have been developed which, in addi- 


tion to that function improve the overall stability of the transmitter. 
Experimental advances have been made in higher powers at the 
higher-frequency television channels. A transmitter of nominal 
power output has been developed suitable for broadcast service. 

In order fully to utilize a television frequency channel, it is de- 
sirable to attenuate most of one picture side-band at the transmitter. 
A method for doing this was evolved and tested in laboratory and 
field with satisfactory results. This was suitable for carrier-frequency 
operation at high power and included constant-resistance circuits 
and phase-correcting networks. Experimental work was also done 
on obtaining the same characteristic at a low level in combination 
with low-level modulation. 

Improvements were made in the mechanical designs and electrical 
characteristics of transmitting antennas. Antennas suitable for 
installation on the small space available on top of tall buildings have 
been designed. The directivity pattern has been improved for 
horizontally polarized antennas so that they have a circular pattern 
in the horizontal plane and directivity toward the horizon in the verti- 
cal plane, resulting in a substantial power gam. The selectivity of 
these structures has been improved so that they have uniform im- 
pedance over more than one 6-megacycle television channel. 

(4) Signal Propagation. Study was given to propagation char- 
acteristics of ultra-short waves in the region of 40 megacycles to 
several hundred megacycles. Comparisons of polarization of the 
radiated wave have been made indicating that a better signal-to- 
ignition-interference ratio and less multipath interference is ob- 
tained with horizontal than with vertical polarization. 

(5) Receivers. Advances were made in television receiver design 
resulting in improved performance and simplification of operation. 
Circuits permitting pre-set station selection have been developed, 
and the number of operating controls has been reduced. The fre- 
quency-band width passed by the receivers has been increased to 
correspond with the increased effective fequency-band made avail- 
able by suppressing one side-band at the transmitter. This results 
in more picture detail. Amplifier tubes of higher transconductance 
have been made available so that more gain and improved signal-to- 
noise ratios can be had, even with the increased band width. Screen 
material for Kinescopes has been developed so that pictures are 
bright and black and white. 

(6) Large Screen Pictures. Progress has been made in circuits 


and cathode-ray tubes for producing large pictures by projection. 
Experimental apparatus of this type has been demonstrated to large 
groups with success. 


Although no new publications appeared during the year, mention 
should be made of the Russian technical journal, Kinomechanik, 
which was in its third volume as a monthly issue. 

The following books of noteworthy interest have been published 
since the last report of the Committee in April, 1938: 

(1) History of Motion Pictures; M. Bardeche and R. Brasillach, translated 
from the French by I. Barry (Norton & Co., New York). 

(2) Photographic Make-Up; W. Meltmar (Pitman Publishing Corp., New 

(3) Der Schmalfilm Tont (Substandard Sound-Film) ; H. Umbehr (Knapp 
Halle (Saale) Germany). 

(4) Color Photography in Practice; D. A. Spencer (Pitman & Sons, London). 

(5) Color Photography for the Amateur; K. Henney (McGraw-Hill Co., 
New York). 

(6) The Eighth Art A Life of Color Photography; V. Keppler (Morrow & 
Co., New York). 

(7) Photography Principles and Practice; C. B. Neblette, 3rd Edition 
(D. Van No strand Co., New York). 

(8) Photographic Chemicals and Solutions; J. I. Crabtree and G. E. Mat- 
thews (American Photographic Publishing Co., Boston, Mass.). 

(9) Processing Miniature Films; P. K. Turner (Link House Publications, 
Ltd., London). 

(10) How to Make Good Movies (Eastman Kodak Co., Rochester, N. Y.). 

(11) Amateur Film-making; G. H. Sewell (Blackie & Son, Ltd., London). 

(12) The American Cinematographer Handbook and Reference Guide; J. J. 
Rose (American Society of Cinematographer f s, Hollywood, Calif.). 

Yearbooks were issued by the following publishers : 

Quigley Publishing Co., New York. 

Film Daily, New York. 

Kinematograph Publications, Ltd., London. 

Photokino- Verlag, Berlin. 

M. Hess, Berlin-Schonberg. 

Abridgements and compilations were issued as follows : 

Abridged Scientific Publications of the Kodak Research Laboratories, Vol. 18 
(1936) and Vol. 19 (1937) (Eastman Kodak Company, Rochester, N. Y.). 

Fortschritte der Photographic (Process of Photography), Vol. 5, Ergebnisse 
der Angewandten Physikalischen Chemie; edited by E. Stenger (Akademische 
Verlags. M. B. H. Leipzig). Reviews progress from 1930 to 1937. 



1 Amer. Cinemat., 19 (Dec., 1938), p. 487. 

2 Internal. Phot., 10 (Oct., 1938), p. 9. 

3 Tech. Cinemat., 9 (July, 1938), p. 91. 

4 Phot. J., 78 (July, 1938), p. 549. 

5 Amer. Cinemat., 19 (July, 1938), p. 270; 19 (Aug., 1938), p. 316; 19 (Sept., 
1938), p. 356; 20 (Feb., 1939), p. 59. 

6 Amer. Cinemat., 20 (Jan., 1939). 

7 Kinemat. Weekly, 261 (Nov. 3, 1938), p. 29. 

8 INMAN, G. E., AND ROBINSON, W. H., JR.: "The Fluorescent Lamp and Its 
Application to Motion Picture Studio Lighting;" to be published in a forth- 
coming issue of the JOURNAL. 

9 Amer. Cinemat., 19 (Dec., 1938), p. 523. 

10 Movie Makers, 13 (Nov., 1938), p. 549. 

11 BAKER, J. O., AND ROBINSON, D. H.: "Modulated High Frequency Re- 
cording as a Means of Determining Conditions for Optimal Processing," /. Soc. 
Mot. Pict. Eng., XXX (Jan., 1938), No. 1, p. 3. 

12 FRAYNE, J. G., AND SILENT, H. C.: "Pushpull Recording with the Light 
Valve," /. Soc. Mot. Pict. Eng., XXXI (July, 1938), No. 1, p. 46. 

13 DIMMICK, G. L., AND SACHTLEBEN, L. T.: "An Ultra- Violet Pushpull 
Recording Optical System for Newsreel Cameras," /. Soc. Mot. Pict. Eng., XXXI 
(July, 1938), No. 1, p. 87. 

14 CRANE, G. R.: "Variable Matte Control (Squeeze Track) for Variable 
Density Recording," J. Soc. Mot. Pict. Eng., XXXI (Nov., 1938), No. 5, p. 531. 

15 AALBERG, J. O., AND STEWART, J. G. : "Application of Non-Linear Volume 
Characteristics to Dialogue Recording," /. Soc. Mot. Pict. Eng., XXXI (Sept., 
1938), No. 3, p. 248. 

16 SCOVILLE, R. R. : "Overload Limiters for the Protection of Modulating 
Devices," /. Soc. Mot. Pict. Eng., XXXI (July, 1938), No. 1, p. 93. 

17 DUSHMAN, S. : "Recent Developments in Gaseous Discharge Lamps," J. 
Soc. Mot. Pict. Eng., XXX (Jan., 1938), No. 1, p. 58. 

18 SCOTT, H. H., AND PACKARD, L. E. : "The Sound Level Meter in the Motion 
Picture Industry," /. Soc. Mot. Pict. Eng., XXX (Apr., 1938), No. 4, p. 458. 

19 Bulletin, Academy of Motion Picture Arts & Sciences (Oct. 10, 1938). 

20 NOEL, E. B., AND FARNHAM, R. E.: "The Water-Cooled Quartz Mercury 
Lamps," J. Soc Mot. Pict. Eng., XXXI (Sept., 1938), p. 221. 



One of the major problems of the motion picture industry in Japan 
in 1938 was the complete ban of foreign pictures, which was made 
effective in September, 1937. This was done by the Finance Depart- 
ment of the Government for purely economic reasons in an effort to 


overcome adverse exchange balances incurred or worsened by the 
China Incident. It is hard to understand why the local producers 
did not find this ban advantageous until one considers that in Japan 
the producers control the major portion of the distribution of both 
local and foreign films. Imported pictures, especially American, as 
has been pointed out in a previous report, are very popular in Japan 
and easily account for their share of admissions. It can be readily 
seen how this source of profit directly assists the Japanese producer 
who happens to be the exhibitor. Furthermore, foreign films are a 
stimulus to the local industry. They furnish suggestions and pat- 
terns to be simulated; they provide the contact between the local 
producer and his western contemporaries. So it is not inconceivable 
that the local producers as well as the foreign distributors were eager 
to see the lifting of the ban. 

Negotiations by the American Motion Picture Association of 
Japan were started in February but it was not until July that an 
agreement was reached and not until October that the ban was finally 
lifted. The agreement was only temporary; however, it allowed the 
importation of approximately 200 pictures before the end of the year, 
and the purchase of a certain amount of exchange in payment for the 
pictures and for the transfer to America of a good share of the pre- 
viously accumulated royalties frozen in Japan. Until the end 
of the year, at which time the agreement was to expire, only about 
100 pictures had been imported, but it is understood that the govern- 
ment has consented to a three months' extension to allow the agree- 
ment to be fulfilled. These newly imported pictures started to make 
their appearance in theaters during the latter prat of November. 
Approximately 40 German pictures were imported into Manchuria 
by the Manchurian Film Monopoly under a German-Manchurian 
Trade Agreement promulgated last year. It was planned that these 
films would also be exhibited in Japan but no adequate scheme had 
been devised for distributing them before the year closed. Several 
Italian films were imported into Japan during 1938. As far as is 
known, only one was exhibited and that with only moderate success. 
It has been suggested that with the Anti-Comintern Cultural agree- 
ments between Japan, Germany, and Italy, exchange of films on a 
barter basis may be favorably considered. Whether or not such a 
plan will be successful will probably be largely determined by the 
amount of active support given it by the governments involved. 
American distributors, while not prepared to meet such competition, 


will no doubt be materially aided by the recognized superior box- 
office success of American pictures. 

Only one piece of governmental legislation was enacted in 1938 
which pertained to the motion picture industry. This law enforced 
from February 1, 1938, limited each theater program to three hours. 
Prior to that time exhibitors screened two foreign features, one 
foreign and one Japanese or two Japanese pictures, plus newsreels 
and shorts, making a program sometimes as long as five hours. The 
new ruling was considered necessary by the authorities to promote the 
public health, to economize on film and to encourage the production of 
fewer but better pictures. This new ruling proved beneficial to 
foreign distributors since it came at a time when they were struggling 
with a shortage of films. There was no serious objection raised 
against the law by the exhibitors. It led, however, in certain in- 
stances to the objectionable practice of cutting pictures, both foreign 
and Japanese, to the point where continuity of the story was de- 

According to the censor's report, as published in the Movie Times, 
there were 560 Japanese pictures of the entertainment variety cen- 
sored during the period from December 26, 1937, to January 10, 1939. 
The actual censored footage amounts to approximately 3,100,000 
feet, not including newreels, whose censored footage is very difficult 
to ascertain. Of these Japanese pictures, 44 per cent are of the modern 
talkie type, 54 per cent are historical or classical talkies, 7 per cent 
are modern and historical silent pictures and 2 per cent are docu- 
mentary or educational. These figures are based on the censored 
footage, which it is felt give the best indication of the activity of the 
studios, since only one print of each picture is censored. The per- 
centage of silent pictures indicated above is open to question since 
other sources indicate that in 1938, as in 1937, the production of 
silent pictures approached 15% of the total. 

On the basis of the above censored footage, the release footage of 
Japanese pictures is estimated between 35 and 40 million feet. The 
release footage of foreign pictures probably did not account for more 
than 10 million feet. Newsreels, whose censored footage is unknown 
but of which many prints are normally made, may have approximated 
a release footage of 5 million feet or more. It is hardly conceivable 
that the total release footage exceeded 55 to 60 million feet since im- 
port restrictions on raw stock limited the local supply to less than this 
figure. Inasmuch as the official censor's report is usually not re- 


leased before October or November of the following year, precise 
figures are well high impossible to obtain at this time. Accepting 
the above estimates, the release footage for 1938 has fallen to about 
80 per cent of that for 1937. Reasons for this decrease may be 
gleaned from what has been mentioned above concerning the ban 
on foreign pictures and from remarks to follow concerning import 
restrictions on raw film. 

In 1937 American pictures accounted for more than two- thirds of 
the foreign pictures released in Japan and during 1938 they seemed 
to maintain this average despite the ban effective throughout the 
first nine months of the year. Of the foreign pictures censored in the 
above mentioned period, American pictures accounted for 75 out of 
a total of 110 or about 68 per cent. British followed with 12, Ger- 
man 11, French 8, and Italian 4. The total censored footage of 
foreign pictures, according to the Movie Times report, amounted to 
some 780,000 feet. 

As was mentioned in the preceding report, European productions 
seem to be steadily gaining ground against American pictures, but 
this does not necessarily imply that they are achieving greater popu- 
larity. Pictures produced under the social and political restraint 
peculiar to European countries are much more likely to find favor in 
the eyes of the Japanese censors, especially in view of governmental 
amity. It must also be borne in mind that the exhibitor is able to 
make more favorable terms for those pictures than for American 

Contrary to expectations, the ban on foreign picture importations 
did not stimulate home production. One reason for this, mentioned 
above, is the producer-exhibitor link peculiar to this market plus the 
decrease in box-office receipts resulting from the ban. A further 
factor, which should not be overlooked, was the shortage of raw film 
stocks. Available figures show that the imports of raw film into 
Japan in 1938 were hardly 15% of the quantity imported in 1937. 
It was beyond the ability of the local manufacturers to suddenly 
meet this tremendous increase in demand. 

The hostilities in China caused a tremendous interest in the pro- 
duction and exhibition of newsreels. In 1937 the censors inspected a 
total of 21,869 prints of newsreels. While corresponding data for 
1938 are not available, it is possible that the number may have reached 
that impressive total although after the fall of Hankow and Canton, 
public interest in newsreels of the China fighting slumped. During 


1937 a considerable number of small theaters were opened for the 
express purpose of showing newsreels interspersed with shorts, each 
theater with a seating capacity of from two to three hundred patrons. 
Early in 1938 it was reported that the total number of theaters show- 
ing newsreels exclusively was 60 but a subsequent report released in 
June placed the figure at 32. 

During 1937 there was considerable activity in the production of 
educational films. Government agencies, newspapers, universities, 
and cultural societies, as well as motion picture companies, partici- 
pated in making a total of 287 such pictures, of which 232 were of the 
sound variety and 55 were silent. According to subject these films 
were: army and navy, 49; tourist, 47; education, 45; industry, 
33; documentary, 30; cartoons, 22; advertising, 19; sports, 9; 
manners and customs, 8; sanitation, 7; amusements, 6; science, 5; 
art, 3; politics, 2; music, 2. Since these films are made primarily for 
private distribution, figures as to lengths of various subjects are 
difficult to obtain. Corresponding data for 1938 are not yet avail- 
able but it is doubtful whether the 1937 mark was equaled due to 
film shortage. 

It is believed that there were no new large theaters constructed 
during 1938 because of the rigid control exercised by the government 
over building materials, especially structural steel. Certainly not 
more than 20 to 30 small buildings with a seating capacity of about 
250 each were built. At the end of 1937 the Department of Home 
Affairs reported a total of 1749 theaters in Japan proper which was 
an increase of 122 over 1936. With three or four exceptions, these 
latter were also predominantly small buildings. It was assumed at 
the end of 1937 that about 85 per cent of all these theaters were 
wired for sound and it is possible that this percentage has been slightly 
increased during 1938. Local made sound systems predominate and 
will continue to do so primarily because importation of such equip- 
ment is prohibited. Of the entire number of sound installations, not 
over 20% are of foreign manufacture. 

The Fuji Photo Film Company, which is practically the only manu- 
facturer of standard motion picture films in Japan, has in the past 
two years carried on an extensive program of expansion in an effort 
to supply the raw film necessary for the local market now that im- 
ported stocks are so severely restricted. Their products include a 
clear base panchromatic negative film, a positive film, and a sound 
recording film. It is also rumored that they will soon market a high- 


speed negative film on an anti-halation support. Although they 
have striven valiantly to overcome the shortage of raw stocks on the 
local market incurred by the import restrictions, it is believed that 
their production capacity is still insufficient to obviate imports en- 
tirely. The normal monthly demand in Japan is approximately six 
million feet of positive film, about one million feet of negative film, 
and about one million feet of sound positive film. 

The use of educational film for teaching purposes is fairly wide- 
spread in Japan, particularly in the colleges and universities. Even 
the so-called middle schools occasionally make use of such film. The 
Department of Education is fully aware of the importance of using 
educational film for teaching purposes and had done much to en- 
courage the domestic production of such films. 

It has been reliably reported that in the small villages and towns 
in the rural sections of Japan, theater owners are becoming interested 
in the possibilities of showing releases on 16-mm films. Although at 
present there are only one or two such theaters, this promises to be 
important in the future. Reasons for this include the decreased cost 
of equipment and pictures as well as the fact that the projection of 
16-mm safety film does not require halls of fireproof construction. 
This venture will not be an immediate success because of the present 
shortage of 16-mm film and also because the motion picture companies 
will have to be won over to the idea of making 16-mm prints. The 
fact that there are two locally manufactured 16-mm sound projectors 
is an encouragement to those interested. 


Summary. American motion pictures continued to enjoy widespread popu- 
larity throughout the world during 1938, although the intensification of difficulties 
abroad has resulted in a drop of 70 to 65 per cent in America's domination of the 
world's motion picture screens. The obstacles encountered have been of diverse 
character, including legislative restrictions, quota systems, high taxes, foreign-ex- 
change controls, occasional excessive censorship, so-called "racial" theories, fervent 
efforts to build up local film industries, active hostilities in the Far East and Spain, 
transfers of territories, and such intangible factors as uncertainty and apprehension. 

Various significant legislative enactments occurred during the year in Europe. 
Great Britain imposed a new quota system, to last for 10 years. Notwithstanding 
the erection of new barriers, American films have continued to enjoy a substantial 
European market. 

During 1938, foreign motion picture production totaled 1706 feature films, against 
1809 in 1937. The countries of the Far and Near East led in production, with 967 
features, as compared with 959 in 1937. Production in Europe fell off sharply, 
the total for all Europe being only 609 features. Latin- American feature-film 
production increased by 40 films to a 1938 total of 130, Mexico being the largest 
producer, with 60 features. 

The Latin- American market at present appears to afford a promising opportunity 
to offset the restriction of our picture markets in other parts of the world. 

Spanish-dialog films have scored notable box-office successes in nearly every Latin- 
American country in which they have been shown, locally produced pictures having 
often exerted a powerful appeal during the past year, because they have portrayed 
familiar aspects of life, in a language understood by the audiences. On the other 
hand, a wealth of recent evidence demonstrates the grave defects and difficulties of 
the motion picture production attempted in certain countries abroad on wholly in- 
sufficient foundations. 

American motion pictures continued to enjoy widespread popu- 
larity in every region of the world throughout the year 1938, although 
there was a decrease in the quantity and value of our film exports. 
As always, our American pictures have won friends during the past 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
March 18, 1939. 

** Motion Picture Division, U. S. Bureau of Foreign and Domestic Com- 
merce, Washington, D. C. 


twelve months through their outstanding merits, but they have been 
confronted by the same impediments and handicaps as in other re- 
cent years sometimes in accentuated form. The intensification of 
difficulties abroad has resulted in a falling off, from 70 per cent to 65 
per cent, in America's domination of the world's motion picture 

The obstacles, which have been of diverse sorts, have more or less 
demoralized the local amusement business. Transfers of territories 
have involved drastic changes in the circumstances governing the 
motion picture trade. The intangible psychological factors of un- 
certainty and apprehension have had an appreciable effect. In many 
cases the spirit of nationalism has been heightened to the disadvan- 
tage of a product such as American films, whose appeal is ordinarily 
designed to transcend geographical boundaries. 

During the past year the ardent fanning of that spirit of national- 
ism has meant, in numerous countries, an ever-increasing fervor and 
energy in the attempt to build up the struggling local film industries 
industries which, despite their obvious faults and feebleness, are apt 
to be supported by governmental action. Often this has meant more 
frequent play dates for locally produced pictures at the expense of our 
American productions. Foreign restrictions on American pictures in 
1938 assumed varied but generally vexatious and embarrassing forms. 
In certain countries, quota systems are entrenched, and that trouble- 
some system shows a tendency to spread. Taxes on motion picture 
business abroad are usually high, and the trend is unmistakably up- 
ward. "Racial" theories and campaigns continue here and there to 
bring difficulties, which are not easy to deal with. Foreign exchange 
controls and censorship were continuing problems during 1938, 
though there have been few startling alterations in those fields. 

It may well be noted, at this point, that certain foreign govern- 
ments have been resorting somewhat oftener, and with more vigorous 
insistence, to the method of diplomatic intervention with neutral 
governments in order to prevent the local showing of American feature 
pictures which authoritarian states choose to regard as objectionable. 
The number of such interventions has not been particularly large 
but the activity, in itself, deserves to be noted. 

Legislative Restrictions. Restrictive measures enforced in Europe 
during 1938 included a new quota system instituted in Great Britain 
on April 1, 1938, lasting for the next ten years. The purpose of this 
new law is to compel distributors and exhibitors to utilize a proportion 

160 N. D. GOLDEN [J. s. M. P. E. 

of British-made films, and foreign producers, including American, 
are induced to produce in British studios quality films for both do- 
mestic and world distribution. 

Italy has placed the distribution of all films, commencing January 1, 
1939, under a Government monopoly. A decree to this effect was 
issued on September 4, 1938, establishing the Ente Nazionale In- 
dustria Cinematografica to purchase and distribute in Italy, its col- 
onies and possessions, all motion pictures imported from abroad. 
Because of the severe terms of this decree and the scant opportunity 
offered for the showing of our American films, all American firms hav- 
ing their own distribution branches in Italy, and those distributing 
films through Italian agents, ceased doing business in Italy on Janu- 
ary 1, 1939. This is something new in restrictions. It is the first 
time that a foreign government has gone into the business of distrib- 
uting motion pictures for the outward purpose of profit. With the 
closing of American branches in Italy, Italian exhibitors will unques- 
tionably feel the severe "pinch" involved in the lack of an assured sup- 
ply of films. At best, Italian films average about half of the normal 
box-office receipts attained by American films, and, in the past, 
Italian audiences have not refrained from expressing their adverse 
reaction to Italian films. 

Germany during 1938 widened its authority and influence in Eu- 
rope by absorbing Austria and the Sudeten territory of Czechoslova- 
kia. Not only did the Anschluss of Austria and the partition of 
Czechoslovakia bring some 1100 additional motion picture theaters 
under the German swastika but also the quota laws of Germany were 
applied virtually shutting out American films. Cooperation and 
"compensation" agreements between Germany and other countries 
also have materially decreased the showing of our American pictures. 
This restrictive tendency is spreading over those countries that are 
looking to Germany for economic assistance. 

On October 12, 1938, a Federal decree went into effect in Switzer- 
land making the importation of motion pictures subject to an import 
permit to be issued by the Federal Department of Interior, which has 
also been empowered to fix import contingents. Up to now, however, 
no quotas have been fixed. It is understood that the primary purpose 
of this new decree is to establish a dependably functioning import 
control of films, which was not possible previously. 

The new agreement of May 18, 1938, effective June 1, 1938, in 
Czechoslovakia permits greater facility of American film distribution. 


It has removed certain threatening restrictions, decreased the cost of 
introducing features and dubbed versions (the recent ceding of the 
Sudeten territory to Germany has diminished the value of the con- 
cessions with respect to films dubbed in the German language), bound 
certain conditions, and established the right of American companies 
to establish their own distribution organizations under equitable 

The Danish Parliament on April 13, 1938, revised the existing mo- 
tion picture law and created a Government distributing agency called 
the Film Central, for the purpose of distributing Danish films that 
are not distributed by the producer himself or by independent Dan- 
ish distributors. The law prohibits the distribution of domestic 
films by local branches or agents of foreign motion picture distribut- 
ing companies. 

Although restrictive barriers in many of the European countries 
are being enforced in greater number, the European market for Ameri- 
can films is far from being lost. Countries such as England and 
France (even with their legislative barriers), Belgium, Denmark, the 
Netherlands, Finland, Norway, Poland, and Sweden, still remain im- 
portant outlets for our American pictures. 

The ban on the importation of American motion pictures into Japan 
was lifted in October, 1938. For a period of 13 months no new Ameri- 
can films were permitted entry, in accordance with the law of Sep- 
tember, 1937, banning luxuries under which motion pictures were 
classified. The new plan permits the entry of 200 American films into 
Japan during 1938, and the transfer of 3,000,000 yen of frozen funds 
in Japan to the United States through the Yokohama Specie Bank at 
San Francisco, where such funds are held without interest for a pe- 
riod of 3 years. Second, it provides for the grant of import and ex- 
change permits for the importation of $30,000 worth of films on the 
basis of a fixed valuation of 1.5 cents per foot. Third, it allows dis- 
tributors to remit these funds on a monthly basis, converted into 
dollars and kept in the Yokohama Specie Bank in San Francisco 
under the same restrictions as noted above for the 3,000,000 yen of 
frozen assets. 

While the lifting of the ban permitted the entrance of our American 
pictures again in Japan, uncertainty is reported to be felt in conse- 
quence of predictions that legislation is being drafted for presentation 
to the Diet, providing for rigid control over all phases of the motion 
picture industry. 

162 N. D. GOLDEN [J. s. M. P. E. 

On Dec. 22, 1938, the Governor of New South Wales, Australia, 
approved an amendment to the Cinematograph Bill of 1938 provid- 
ing for a Theater and Films Commission to replace the Film Advisory 
Committee, and setting up new provisions of the Quota Act. The 
new act provides for a 15 per cent quota on the part of exhibitors for 
British quota films, and a 2 l / 2 per cent quota on the part of exhibitors 
for Australian pictures. Exhibitors, under the new act, have a maxi- 
mum rejection right of 25 per cent plus 2*/2 per cent for Australian 
films, or a total of 27 Va per cent. 

Latin- America. The Latin-American market at present appears 
to be the market our American distributors are seeking to offset the 
restricted markets in other parts of the world. With 5239 potential 
theater outlets, and with new theater construction increasing each 
year, American companies are coming to the realization that here is a 
geographical area that should receive closer attention. Economically 
it would be unwise for our companies to encourage production in 
South American countries; however, American companies should 
produce in Hollywood Spanish-dialog films employing stage favor- 
ites brought from South America and placed in a Hollywood setting, 
with the use of reconstructed sets and Hollywood technic. In this 
manner, production costs can be kept at a minimum, and producers 
will have Spanish-language films available to carry then- other Ameri- 
can product which is now being frequently shoved into the back- 
ground by Spanish-speaking productions from Mexico and the Ar- 
gentine. A case in point is the drop in the showing of American films 
in Peru from 70 per cent of the total in 1937 to 49 per cent in 1938, 
which is attributed primarily to the augmented number of Spanish- 
dialog pictures from Mexico, Argentina, and Peru, itself. Although 
none of these films approached the quality and standard of our Ameri- 
can films, they helped to consume playing time that might otherwise 
have been obtained by American films. 

The one important market in Latin-America which may become 
beset with difficulties is Argentina. In the closing sessions of the 
Argentine Congress in 1938, a bill was introduced which would give 
definite powers of regulation and control ot motion pictures to the 
Institute Cinematografico Argentine. This bill would encourage a 
national industry through prizes, subsidies, and other means; it 
would establish a central censorship board, with Argentine films 
exempt from censorship taxes; it would exempt import duties on 
raw film and equipment accessories necessary in the production of 


Argentine films; it would encourage the future establishment of a 
Government film studio. While the bill will not be acted upon until 
the next session of the Argentine Congress, it gives a clear indication 
which way the trend is moving, and may be the forerunner of a 
quota law. Recently in this market a ban has been placed upon the 
importation of film advertising matter. This ban forces our American 
companies in the future to print their posters and lithograph material 
in Argentina to aid the native printing and lithographing trade. 

As an indication of the rising legislation designed to hamper the 
distribution of our American films, one finds that during 1938 a new 
tax was levied against American distributors in Guatemala. For- 
merly, distributors paid $300 per year for the right of operation. This 
has now been changed to 10 per cent of the distributors' share of 
entry receipts. A deduction of 25 per cent from the distributors' 
share may be made to cover expenses, and the 10 per cent is applied 
to the balance. 

During the year 1938, Cuba attempted to pass an exhibitors' 
quota providing that for every 7 imported films one Cuban picture 
must be shown. The House of Representatives failed to pass this 
quota. When one realizes that only two Cuban films were produced 
during 1938, the impracticability of the proposed legislation is clearly 

American Film Exports. Statistics of American motion picture 
film exports for the full 12 months of 1938 show a 13,000,000-foot 
decrease in positive and negative sound and silent films as compared 
with those exported during 1937. During the year 1938 a total of 
202,526,821 feet of motion picture entertainment films, both sound 
and silent, with a declared value of $4,519,594, were exported to all 
foreign markets, as compared with 215,721,956 feet of film, with a 
value $4,797,641, for the year 1937. This is a decrease of 6.1 per 
cent during the year 1938 in the total of all films exported. 

The valuation of these exports merely represents the declared 
value of the raw film cost at 2 to 3 cents per foot and does not repre- 
sent the true value in dollars received by American distributors for 

The foreign-exchange situation has not changed perceptibly during 
1938. In 57 out of 94 countries throughout the world, up to Decem- 
ber 31, 1938, foreign exchange was liberal and easy, ranging from 
prompt to retarded payment. In the remaining 37 countries, foreign 

164 N. D. GOLDEN [j. s. M. P. E. 

exchange is "tight and normal" in 9 countries, "tight and slow" in 
15, and "restricted" in 13 countries. 

Censorship. Censorship abroad has not been unusually trouble- 
some during 1938. In most cases, the censorship rules have been 
carried out in a spirit of reasonableness, moderation, and judicious 
liberalism. In certain instances, though, it would seem that the literal 
terms of the local censorship regulations may be regarded as unduly 
stringent. For example, we learn of films being banned occasionally 
on such grounds as mere "banality" or "the conveying of prejudicial 
illusions." In Greece, pictures portraying stories of the French 
Revolution are taboo, because, though the events happened a century 
and a half ago, they are "connected with political or social movements 
of revolt." Great Britain forbids all films that it considers "horrific," 
and thus the picturization of certain novels written by that nation's 
own distinguished authors are excluded from its screens. "Any- 
thing that might offend local sentiment" is the broad and vague 
description of the sort of dramatic themes that might mean "thumbs 
down" in Germany. In India, one is forbidden to show, in motion 
pictures, "organized knuckle fights" or "profuse bleeding" and 
the same country also bars scenes depicting "relations of capital 
and labor," a rather onerous restriction on producers interested in 
showing the vivid drama inherent in economic problems of the 
present day. One is inclined to agree with a report from British 
Malaya, expressing the opinion that apparently "the local censor 
does not take into consideration the growing sophistication of native 

Foreign Film Production. During the year 1938 foreign motion 
picture production totaled approximately 1706 feature films, as 
against 1809 features in 1937. The countries of the Far and Near 
East led in production, with a total of 967 features in 1938, as com- 
pared with 959 in 1937. Japan with 575 features was again the lead- 
ing producer, followed by India with 200 features. The Philippine 
Islands account for 67 films, Hong Kong 53, China 33, Egypt 16, 
Siam 10, Australia 8, Chosen and Formosa 2 each, and New Zealand 1. 

Production in Europe fell off sharply in 1938, primarily as a result 
of England's decline in production. For the year 1938, all countries 
of Europe produced a total of 609 features. The following countries 
were producers of feature films during 1938: Germany 137, France 
122, England 85, Russia 51, Italy 47, Czechoslovakia 41, Sweden 30, 
Hungary 26, Poland 25, Finland 20, Denmark 9, Norway 4, Turkey, 


Belgium, and Portugal 3 each, Netherlands 2, and Switzerland 1. 

Latin- American feature-film production during 1938 took an up- 
ward jump of 40 films. During 1938, 130 full-length features were 
produced as compared with 90 in 1937. Mexico was the largest 
producer, turning out 60 features. Argentina jumped its production 
to 50 features from 30 in 1937. Peru increased from 2 in 1937 to 11 
in 1938. Brazil produced 4, Cuba and Uruguay 2 each, and Vene- 
zuela 1. 

Increased production in Latin-America substantiates the fact of 
the Latin- Americans' desire to see and hear their own stars speaking 
their native tongue. In many markets these native pictures, regard- 
less of their quality, have far outdistanced some of our biggest 
American productions. This especially is true in the rural communi- 
ties of Latin- America. Films produced in Mexico and Argentina 
have scored notable box-office successes in nearly every Latin-Ameri- 
can country in which they have been shown. This is true also of 
those Spanish- dialog films produced in Hollywood employing actors 
brought from Latin-American countries, speaking Spanish, and 
placed in stories having a Latin-American atmosphere. 

It is keenly interesting to note some of the salient characteristics 
of "infant" picture-producing industries in various foreign countries 
during the past year, as recorded by competent observers on the 
spot. On the one hand (as just indicated) we must frankly recognize 
the fact that the locally produced pictures often exert a powerful, 
though naturally restricted, appeal ; they do so not only because they 
speak the language of the people but also because they "portray 
outstanding aspects of the national life, showing typical landscapes, 
dances, and music" (to paraphrase one report from the Caribbean 
region) . One picture in particular, produced south of the Rio Grande, 
is noteworthy as having achieved a really "tremendous hit" during 
1938 in a number of Latin- American countries where the people 
welcomed it eagerly by reason of their perfect comprehension of the 
language and the animating moods of the action. 

On the other hand, we have a wealth of recent evidence to demon- 
strate the grave defects, shortcomings, and often insuperable diffi- 
culties of the motion picture production attempted in certain coun- 
tries, very largely on the basis of nationalistic ambition, without a 
solid foundation. From one European country we hear of complaints 
by audiences that ' ' the local pictures seem ' stationary. ' ' ' Lacking the 
proper facilities for moving the cameras, or for creating a fascinating 

166 N. D. GOLDEN [j. s. M. P. E. 

variety of settings, the local producers are unable to avoid, in their 
films, an annoying and exceedingly tiring "static" quality. Again, 
we learn that "the make-up of even the leading players in local 
pictures is often far from flattering," and the spectators are thereby 
repelled or amused in the wrong places. (What is being said here 
does not refer, of course, to major producing countries such as 
France and Great Britain.) 

"The lighting is generally hard and flat" is a criticism voiced of the 
pictures being turned out in one Far Eastern country in vivid 
contrast to the magical lighting effects achieved in nearly every 
American picture. The music used in some foreign studios is that of 
old, imported phonograph records and, when the picture is edited, 
there occurs from time to time an abrupt and disconcerting "chopping 
off" of the musical background, perhaps in the middle of a phrase. 

"Direction is deficient," is the 1938 dictum from another part of 
the world, with respect to locally produced pictures. A basic handi- 
cap noted is that "many local films have been and are being produced 
'on a shoestring.' ' "Technic lags behind" is the point emphasized 
in still another discussion of foreign production, and, with respect to 
one major country, it is asserted today that the local producers "can 
not provide the spice and variety characteristic of American pictures." 

In one foreign country, with a vast population, the motion picture 
production attempted by various local interests has, in general, 
"gained the reputation of being a poor financial risk and a highly 
speculative venture." In another part of the world, the local pro- 
ducers are described as being "unable to do any really serious work," 
the existing companies being "small, poorly organized, and inade- 
quately financed." In one very substantial nation of western Europe, 
we find that "the high percentage of financial failures registered by 
local motion picture productions discourages fresh investments." 

In northern Europe, one splendid new theater took the rather 
staggering loss of 2000 crowns daily while showing a locally produced 
picture, of which high hopes had been entertained. In that country, 
we ascertain, producers have experienced great difficulty in engaging 
satisfactory casts; one reason for this is the fact that they are com- 
pelled to rely exclusively on stage actors and actresses, and these 
artists can act in pictures only during the three months of their summer 
holidays. In one British Dominion, the year 1938 witnessed the 
production of one local feature which was "shown in one theater 
and immediately forgotten." 


In one of the European countries, in 1938, the sum of 600,000 
francs was sunk in the production of a local film which, when it was 
finished, found no market whatever. It is said that the film can not 
be sold and that the money invested in it is likely to represent a 
total loss. A number of firms, in the same country, are attempting 
to produce pictures in a structure converted from other purposes, 
where they have a working-room only 8 meters wide by 20 meters 

Summation. From such reports as these, covering 1938, one 
gets a clear idea of the flaws, misfortunes, frustrations, and frequent 
unfavorable reactions involved in the attempt by various foreign 
interests to establish local motion picture production abroad where 
there is actually scant necessity or justification for the effort. In 
numerous cases the difficulties seem well nigh insurmountable, since 
they arise out of the inherent limitations of the country. In other 
instances, of course, the present handicaps are to some degree tem- 
porary, and faults will be gradually corrected as circumstances are 
altered and development proceeds. 

But whether the future, in a given country, presents one prospect 
or the other, today the fact is indisputable that a very considerable 
proportion of the motion picture production abroad is of a quality 
markedly inferior to prevailing American standards. 

In view of that fact, what action should be taken by our American 
companies in order to maintain a position of superiority over their 
competitors in the markets of the world? To what major measure 
can they today resort, with the object of checking trends which we 
must acknowledge to be adverse? What dynamic attraction or 
allurement can be exerted, of greater potency than the local appeal 
of a spectator's mother-tongue and his natural fondness for familiar 
scenes and ways of life ? 

Plainly, before all else, we must emphasize to the utmost the 
contrast in quality between our good American pictures and the typi- 
cal product of local producing industries abroad. We must make 
that contrast as vivid, as striking, as impressive, as it can possibly be 
made. Persistently and adroitly, we must make the foreign movie- 
goer acutely conscious that the American picture is a product of 
decidedly superior quality of rich and varied artistry, of entertain- 
ment value unmatchable in the run-of-the-mine output of our com- 
petitors abroad. We must make this "high-quality" factor so uni- 
versally recognized that local audiences abroad will have no desire to 

168 N. D. GOLDEN [j. s. M. p. E. 

see inferior films that owe their existence simply to some government 
legislation or subsidy. 

Very recent news dispatches tell us how certain foreign audiences, 
deprived of American pictures, have manifested their displeasure in 
the strongest possible terms. 

It is unwise for us to try to export mediocre films. Foreign audi- 
ences, in numerous countries, get an abundance of that kind of pic- 
tures from their own studios. American distributors should send only 
their choice Grade A films to the foreign market. If the choice is 
between our B type of films and a picture from a native studio, the 
latter (one need hardly say) is almost invariably preferred. An 
examination, over the past few years, of our best revenue-producers 
in the foreign market discloses that those films listed among the best 
pictures shown here in the United States have also been the biggest 
revenue-remitters . 

Certainly, therefore, it behooves our American companies today, 
more than ever before, to make a very careful selection of the films 
to be exported and not to send, more or less at random, pictures 
that might affect unfavorably the ultimate gross returns, while 
impairing our indispensable prestige and reputation for superior 

As we advance into the new year 1939, the factors to be relied 
upon, in maintaining our position in foreign markets, may still be 
defined as the simple, basic elements of our unmatched scientific 
skill in motion picture production our amazing capacity for devising 
new and really wonderful methods our determination to achieve 
artistic and enthralling camera-effects the incomparable richness of 
our material facilities and resources and our unequaled variety and 
range of every type of acting talent. Together, these things spell 
quality and it is quality that will continue to attract foreign audi- 
ences to American pictures. 


MR. CRABTREE: With regard to foreign versions, what percentage are pro- 
duced with foreign actors; what percentage have superimposed titles; and what 
percentage are "dubbed," with an attempt to synchronize the lip movements 
with the foreign language? 

MR. GOLDEN: In certain markets of Latin- America our pictures are preferred 
in their original versions with superimposed titles, but in other markets we can get 
away with dubbed versions. A great number of Latin- Americans are trying to 
improve their knowledge of English, and the use of motion pictures as an educa- 
tional aid is very valuable. Dubbing is done in certain of the European coun- 


tries, particularly in Paris and London. In other markets a great deal of super- 
imposing is done. We send our pictures in the original version and superimposing 
of the titles is done where facilities are available. 

MR. CRABTREE: Mr. Sponable once showed us a very interesting example of 
synchronization of lip movement in the English version with a foreign language. 
Perhaps he can tell us whether this business is growing or diminishing? 

MR. SPONABLE: I think it is about stationary. As Mr. Golden has said, we 
dubb a large number of our pictures abroad in French, Italian, and Spanish. 

MR. CRABTREE: I suppose the reason for this is that when separated from the 
source, one loses the colloquialisms of a language. 

MR. RYDER: Relative to the number of foreign dubbed pictures, it might be 
interesting to know that in the past Paramount has dubbed about 60 per cent of 
its pictures. This means that French or Spanish lines are used to replace the 
original English dialog. The recent foreign situation, I feel sure, has changed the 
percentage and the uncertainties that now exist will cause a further change so 
that the percentage will be somewhat different from what it has been in the past. 

MR. ERMOLIEFF: I think every American producer and distributor adapts 
as many pictures as he can. If a picture is some sort of success we try to adapt 
it because the greater masses care to see the picture in their own language. 

In Snow White we had a picture that made tremendous money abroad. It was 
dubbed in about ten languages, even in the languages of small countries, in which 
we would not dubb other pictures. Pictures that we dubb in French are released 
in Belgium and other French-speaking countries; we dubb in Italian for all the 
Italian-speaking countries; and for the Scandinavian-speaking countries we 
usually dubb in Swedish. 

MR. CRABTREE: With what degree of success is synchronization of the lip 
movement accomplished or do you let the foreigners worry about that? 

MR. GOLDEN: They worry very much about it. You can not synchronize 
quite perfectly, and there is always some sort of difficulty. We do not have as 
much trouble with German, because the German lip movements are more similar 
to ours. We have more trouble with French, and still more with Spanish, de- 
pending more or less upon the picture. 

MR. LESHING: Mr. Crabtree would like to know how successful we are in 
synchronizing the lip movements. We do it not only in the foreign versions but 
in our own productions, when we give the voice of a singer to someone else who 
can not sing. I do not know how much of the dubbing is being done here. We 
do, however, supply material to our foreign department for practically every 
country that uses our productions, and they do their own dubbing. Whether 
they are doing it in Italy, France, or Spain we do not know, but we do supply them 
with material for the dubbing which a couple or three years ago was done here in 
the United States. 

MR. GOLDEN: This dubbing is one of the worse worries that the industry has 
to face in the foreign market. The quota laws or legislative barriers that exist in 
some of the countries require that the dubbing be done in the country in which the 
picture is to be shown. France and Italy have definite laws in this regard, as 
well as some of the Scandinavian countries, and these provisions are designed 
primarily to keep the local struggling industries going. 

MR. CRABTREE: As in the example that Mr. Leshing mentioned, there can be 

170 N. D. GOLDEN 

a great amount of effort put into studying the lip movement and then changing 
the foreign dialog so as to synchronize with these movements. Are the foreigners 
putting any degree of concentrated effort into the dubbing or do they just make 
the lines of the dialog fit the time of projection? 

MR. GOLDEN: The dialog must fit the time of projection. In France it is 
required that something be said in the French titles that the film was dubbed in 
France and that the actors are not speaking the French language. That is a part 
of the law. Some eight years ago I had the privilege of seeing at the M-G-M 
Studio some of their German-dubbed pictures. Mrs. Golden who is quite a 
student of German was with me. We looked at a picture starring Marie Dressier 
and Wallace Beery and after about ten minutes had passed Mrs. Golden said to 
me, "I never knew Marie Dressier or Wallace Beery spoke German." That is 
how well it was done then, and I imagine the improvements today far exceed the 
accomplishments of eight years ago. 

DR. FRAYNE : In sending the sound effects and the music on these pictures for 
dubbing are those sent on duplicate positive material, negative, or what? 

MR. RYDER: In the case of Paramount our dubbing (re-recording) procedure 
is as follows: On each sequence or section of the picture to be dubbed, we first 
make a take for our domestic release, using the English dialog. Subsequently, 
for foreign dubbing we make a second take on the same sequence, omitting the 
English dialog, but incorporating all the music and effects used in the original 
take, plus such added effects as are necessary to give continuity of sound after the 
dialog has been removed. Prints from the negative thus obtained are sent to the 
foreign country, where the foreign dialog is recorded and composited by re-record- 
ing with the track that we supplied, thus obtaining a new negative for foreign 
release. Incidentally, the prints that we send to foreign countries are printed 
heads and tails so as to save film footage. 

DR. FRAYNE: In foreign countries they are favored in getting an original re- 
cording of their own, as they then do not have to go through a re-recording 
process. I presume in case of negatives made in foreign countries the final print 
is made frequently from this negative without going through a re-recording 
process resulting in loss of quality usually entailed in this process. 

MR. RYDER : The sound prints that we supply to the foreign market, for in- 
stance to Paris, have effects and music but no dialog. The Paris studios syn- 
chronize the dialog to the picture action and substitute French dialog for the 
English. Then, by the dubbing process they put the two together and obtain a 
new composite negative with French dialog and the re-recording effects. That 
new composite negative then is used for release printing in French. 

DR. FRAYNE: This might still be better than a photographic duplicate. If 
they took a foreign language negative made here, and made a photographic dupe 
over there, the quality might be inferior to what they are getting now in their 
present dubbing scheme. 

MR. RYDER: That is correct. Further, from a commercial standpoint, I 
believe it pays them to do as much work as possible in the foreign country. 




Summary. During the past year a special triple-head projection mechanism 
has been in use at the Paramount Studio for transparency process work. This 
device permits the projection of three separate background positives, through three 
separate projection heads, superimposing the images upon a single screen to form a 
single registered image of greater size and brightness than would otherwise be possible. 

The general construction, use, and applications of the device are described. 

The projected background or transparency process of composite 
cinematography, has come into its present widespread use not merely 
because it permits the accomplishment of scenes which would be im- 
possible or dangerous to film by conventional methods, but because 
it makes possible substantial savings in time, effort, and money. 
To send a major-studio production unit on an extended location trip 
can, under modern production conditions, be fantastically expensive ; 
such units have been known to cost their studios five and ten thou- 
sand dollars per day, and to spend weeks at a time on location, wait- 
ing for favorable weather. 

If a comparable location result can be achieved in the studio, 
with the only added expense involved being that of sending a skeleton 
transparency background camera crew, with perhaps an assistant 
director and such supernumerary players as doubles, for the principal 
actors, extras, stunt riders, and the like, to the location, and there- 
after completing the scenes with the principals in the studio, it is 
obvious that worth-while economies must result. 

But if such a method or process is to achieve these economies, 
two basic considerations must be fulfilled. First, and most impor- 
tant, the overall screen effect of the composite picture must be 
real and convincing in its appearance and it must preserve the illu- 
sion that both the foreground action and the background were filmed 
together, at the same time and place. Second, the physical scope of 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 21, 1939. 

** Paramount Studios, Hollywood, Calif. 


172 A. F. EDOUART [j. s. M. P. E 

the process must be such as to place a 'minimum of technical restric- 
tions or better, if possible, no restrictions at all upon the creative 
efforts of the director and writers. 

Modern methods, materials, and equipment are enabling modern 
transparency cinematographers to meet these requirements with 
increasing success. The detailed story of the progressive steps of 
these various methods and their evolution has been related so fre- 
quently at previous conventions of this Society, and published in the 
JOURNAL, that it seems unnecessary to go into that detail at this 
time or to attempt to retrace this familiar ground. 

It is significant to note that the most highly important trans- 
parency development made during the past year is that of the triple- 
head projection mechanism. This device permits the projection of 
three separate background positives, through three essentially 
separate projection heads, superimposed upon a single screen to 
form a single registered image of greater size and brightness than 
would otherwise be possible. Such devices were developed inde- 
pendently and practically simultaneously, unknown to each other 
by the Transparency Department of the Paramount Studio and the 
Special Process staff of Warner Brothers-First National Studios. 
The Paramount device will be described here. 

The triple-projection head was the outgrowth of the acknowledged 
difficulties of securing sufficient screen brightness for extra large 
screens for black-and-white work and also for an added exposure 
level when working with three-color systems of natural-color cine- 

The first experiments were along the lines of utilizing in reverse 
the same idea in projecting, as Paramount had in the dual screen 
camera set-up; for photographing original background plates. In 
other words, a double projector with superimposed images to secure 
more scope on larger black-and-white scenes. This idea almost 
immediately expanded, however, into the present triple-head idea. 

At the same time we were invested with the problem of securing 
greater scope in color transparencies. Bearing in mind the economic 
angle of requiring three color plates to project at a time, the first 
experiments were along the logical lines of additive projection, using 
black-and-white background plates with appropriate tricolor filters. 
This method proved unsatisfactory due mainly to the great loss of 
light, attributable to the transmission factor of the filters, so we 
adopted the use of Technicolor sub tractive full color plates. The 

(Courtesy of Paramount Pictures, Inc.) 

FIG. 1. (Upper.) Front view, showing the triple pro- 
jection head layout. The left and right lens images are re- 
flections from the left and right heads, respectively. 

FIG. 2. (Center.) Front view of the three heads, mir- 
ror assembly, the right and left-hand lens remote focus 
motors, and synchronous drive motors. 

FIG. 3. (Lower.) Front view of triple projection lenses 
through silencing funnel. The right and left lens images 
are reflections from the two first surfaced reflecting mirrors. 

174 A. F. EDOUART [j. s. M. P. E. 

ultimate resulting method employs in' color three full-color positive 
plates and has proved most satisfactory for color transparencies. 

The projection unit consists of three fundamentally standard 
transparency projection units, each with its own lamp house, move- 
ment, and optical system, all three sharing a common base (Fig. 1). 

One of the three heads the one located in the center is mounted 
in the conventional position, parallel to the axis of projection. The 
other two are mounted at right angles to the axis of projection, 
facing respectively inward to that axis. These two outer heads 
project their images onto the screen by means of first surfaced alumi- 
nized mirrors reflecting their images at an angle of slightly more than 
45 degrees to superimpose themselves over that of the directly pro- 
jected center lens into a single registered triple image. All three 
projection mechanisms are driven by electrically interlocked syn- 
chronized motors. The lenses are focused separately by a reversible 
type motor operated remotely from the camera. 

The two outer lenses are mounted in the same horizontal axis 
plane with the center lens, having their respective reflecting mirrors 
mounted in front of the lenses, allowing sufficient space between the 
mirrors to permit the image from the center lens to pass between 
them (Fig. 2). All lining up work is done with the center machine 
first, then each side machine is focused and aligned up separately to 
match that of the center machine. When completed, all three 
images are superimposed into perfect registration as a single image 
of beautiful photographic quality, of smoothness, of light steadiness, 
and of flatness of field. 

Parallax is adequately compensated for by independent horizontal 
and lateral movement of each of the outer lenses, in a manner directly 
comparable to the sliding front-board of a still camera. The outer 
lens images are adjusted individually, to coincide perfectly with 
that of the center lens and each other (Fig. 3). The reflecting mirrors 
have micrometer lateral and horizontal adjustments to bring the 
images of any focal length lenses into coincidence with each other at 
the screen. 

The base supporting all three heads may be pivoted horizontally 
or tilted up or down vertically, moving all three heads, lenses, mir- 
rors, and lamp houses as a unit (Fig. 4). After the center image is 
in proper perspective and focused for the shot, it does not generally 
take longer than from three to seven minutes to register perfectly 
the other two images. 



(Courtesy of Paramount Pictures, Inc.) 

FIG. 4. (Upper.) Right side view looking forward, 
showing center and right-hand equipments. The right- 
hand head is a left-hand thread up. 

FIG. 5. (Center.) Center and right-hand projection 
heads. The right-hand head is a left-hand thread up. 
FIG. 6. (Lower.) Left side looking forward. 

176 A. F. EDOUART [j. s. M. P. E. 

The use of the triple-projection device naturally increases screen 
illumination enormously. A conservative figure for this increase, 
which may be varied by any of several controllable factors, is approxi- 
mately 280 per cent more than is possible with any similar single 
projection equipment. 

This inevitably means that a large picture area may be used. 
But in addition, it gives several added advantages in flexibility and 
cumulative quality, as mentioned above. 

With the increased illumination, when used under normal cir- 
cumstances, darker prints may be used than heretofore, thereby 
obtaining a wider gradational scale, and as a result improved screen 
quality. In the same way, the triple-head provides a vastly greater 
reserve of illuminating power, and it is possible to utilize this in- 
creased light output to secure finer quality results rather than be 
forced to strain single-unit illuminants of the same type to the maxi- 
mum, and be required to use lighter prints. 

In the same way, it is possible if desired on smaller picture areas, 
to stop down fast projection lenses to assure maximum definition 
in the projected picture. 

The increased screen brilliance naturally makes it possible in 
many cases also to stop down the lens of the composite camera and 
thus obtain a desirable increase in depth of focus. 

Graininess in the projected picture has always been a difficulty 
in the transparency process. The triple-projection method, as can 
be readily seen, will obviously lessen this problem. The super- 
position of three separate prints necessarily tends to minimize the 
visual effect of grain by one-third, since the individual grain-images 
tend to overlap and cancel each other out. The same is true with 
other print imperfections. 

The use of three light-sources likewise tends to minimize flicker 
and light fluctuations caused by variations in the intensity of the 
projection lamp. It is extremely unlikely that the three arcs could 
ever flicker synchronously. Therefore, when compared to the effects 
of arc flicker in a single projector, a comparable flicker in any of the 
three arcs of the triple projector would result in only a 33 Va per cent 
variation on the screen, instead of 100 per cent variation, as in the 
case of a single light-source. 

A flatter field of illumination naturally results. Since so great a 
quantity of light is available for normal shots, the light-source focus 
does not have to be concentrated to so intense a spot on the aper- 

Aug., 1939] 



(Courtesy of Paramount Pictures, Inc.) 

FIG. 7. ( Upper.) Left side of the transparency projection equipment 
as mounted in booth. 

FIG. 8. (Lower.) Rear of booth showing, rewind table and film cabi- 
nets. The center cabinet is for air conditioning. 

178 A. F. EDOUART [j. s. M. p. E. 

tures, thereby giving a more even illumination, and in addition helps 
smooth out the center hot-spot so common with single projection. 
This method also has the additional advantage of tending to lessen 
the heat falling on each of the prints, resulting in their longer useful 

In addition to all these logical duping advantages, the triple- 
projection system gives added light-control possibilities. It is, for 
instance, entirely possible to vary the intensities of each of the 
three arcs, balancing the overall intensity to the requirements of the 
scene in hand. 

Furthermore, it is by no means necessary, nor even always de- 
sirable, to utilize three prints of identical densities in this system. 
It is quite possible (and very frequently done), for example, to use 
one light print and two darker ones, or one dark and two light ones ; 
or prints of three entirely different densities. This gives a great 
range of control not alone of screen brightness, but of contrast, 
gradation, and shadow illumination in the projected picture. 

To summarize, the triple-projection system, through increasing the 
available illuminating power approximately three-fold, not only 
increases the scope of the transparency process by making possible 
the use of larger background screens, but improves the quality ob- 
tainable on transparency scenes of normal scope by yielding increased 
gradation al scales and definition and minimizing the undesirable 
effects of graininess print imperfections and light-source fluctua- 

One comment, however, is now necessary in connection with fur- 
ther consideration of the transparency process problem. It must 
always be kept in mind that the composite picture resulting is the 
effect of a blend of two photographically contradictory elements. 
The foreground action is an original photograph, consisting as it 
does of images of action taking place directly before the composite 
camera. The background, on the other hand, is essentially a "dupe," 
consisting of re-photographed images of the scene projected upon the 
background screen. 

The foreground portion naturally partakes of all the favorable 
qualities of definition, gradation, etc., of any well photographed origi- 
nal. The background must be prevented from partaking of the 
equally familiar, unfavorable characteristics of a dupe, including 
impaired definition, gradation, and exposure values, as well as all 
other imperfections, such as scratches, digs, abrasions, development 

Aug., 1939] 



(Courtesy of Paramount Pictures, Inc.) 

FIG. 9. (Upper.) Front and left side of the 
transparency triple projection booth. 

FIG. 10. (Lower.) Back of the booth, showing 
doors and service outlet connections. 

fluctuation, etc. If the composite scene is to be convincing, the 
entire scene must maintain uniform photographic quality, visual 
perspective, and mechanical registration. This can only be ob- 
tained through coordinated perfection of a long chain of individually 
small details extending through every step from the exposing of the 
original background negative to the processing of the ultimate com- 

180 A. F. EDOUART [j. s. M. P. E. 

posite negative. Errors in any of these details are extremely likely 
to be cumulative and to build up to an overall error of such magnitude 
as to seriously jeopardize the illusion, or even spoil the shot. 

It has already been pointed out that the dramatic and economic 
usefulness of the process is dependent upon the physical scope of the 
process being sufficient to allow the director freedom closely com- 
parable to what he would enjoy if his company was working upon the 
actual location. It is of very little use to have a process that can 
put Gary Cooper in Paris, or Barbara Stanwyck in Wyoming, if 
such scenes must be restricted to close shots of one or two players, or 
if the movements of the actors must be restricted. 

When the transparency process was first introduced, we were 
happy when we could use background screens measuring six or eight 
feet wide, and perhaps make scenes showing our foreground actors 
from head to ankles. As both the technic of the process and the 
equipment and materials for it improved, screens ten, twelve, and 
fourteen feet in width came into use. When, about a year ago, we 
found it possible to use screens twenty and twenty-four feet wide, we 
felt that we had achieved very nearly the ultimate. Today, with 
the development of the triple-head machine, we are able to use thirty- 
six foot screens for black-and-white work, and look forward con- 
fidently to the time when improved equipment now on the drafting- 
boards will enable us to use screens fifty feet or more in width! 

A very important contributing factor in this advance of quality 
improvement has been the relatively recent introduction of the 
modern high-speed films. To the production cinematographer, 
the introduction of these films has meant primarily an opportunity 
to reduce illumination levels. To the transparency cinematographer, 
however, the same development meant quite a different application 
of this extra speed one which means an even greater asset than 
using less illumination. 

One of the greatest problems of transparency process camera- 
work has always been that of keeping the actors and the projected 
background both in adequate focus. In close shots, where screen 
and actor are close to each other, this is relatively simple, even when 
using the lens at or near maximum aperture. But as screen-size 
and the physical scope of transparency shots increase, this problem 
becomes more serious, as the physical separation between actor and 
screen increases. 

The greatest benefit derived from the new films, from the trans- 

Aug., 1939] 



parency standpoint, is therefore not as might be expected, simply the 
added speed, making it possible to utilize the lower illumination- 
levels that would follow; but instead, profiting by the added speed 
to a greater measure by stopping down the lens of the composite 
picture, thereby obtaining considerable increased depth of focus. 

How greatly this factor helps us is best brought out by referring 
to some of the transparency scenes currently being made for Para- 
mount's production Geronimo. In some of these scenes, the set 
represents the encampment of an army detachment being attacked 

(Courtesy of Paramount Pictures, Inc.) 

FIG. 11. Layout of projection booth. 

by Geronimo's Apaches. In the long shots, dual projectors and 
screens are used, the composite camera and background projectors 
are at the extreme opposite ends of the big stage a separation of 195 
feet. The screens are at times as much as 70 feet or more distance 
from the camera, while the distance from the actor nearest the camera 
is not more than 12 to 18 feet. 

Obviously, if the shot is to be convincing the actors as well as 
the intermediate set and the background screen's image must all be 
in adequately sharp focus. In addition, during some of these scenes, 



[J. S. M. P. E. 



(Courtesy of Paramount Pictures, Inc.) 

FIG. 12. Details of equipment. 


40 to 50 horses and their riders, representing the attacking Indians 
or the rescuing cavalry, must ride through the set at various points, 
sometimes quite close to the screen, while important foreground action 
is taking place. 

It is clearly an optical impossibility for any lens of normal focal 
length, used at the maximum aperture of //2.3; and focused on the 
most important foreground or intermediate plane action, or even 
with a split focus, to carry adequate depth of focus to cover the over- 
all range, which is virtually from 12 to 18 feet to four or five times 
this foreground distance. However, with the present fast negative 
film, which permits stopping the lens down to a//3.5, or a sometimes 
smaller aperture, such an achievement becomes possible. It is not 
too much to say that without the aid of these new modern fast 
films, the dual screen or other type large-scale transparency scenes 
could not be nearly so well accomplished. 

Under modern present-day economic conditions, it is likely that 
many productions would not, or could not, be made for any cost 
even remotely permitting hope of a profit, were it, not for the ad- 
vancements made in film manufacture, combined with the use of the 
advanced scope of the transparency and other special processes. 

In conclusion, the writer wishes to stress one vital factor con- 
cerning the employment of the transparency method. In its most 
accustomed applications, it is not intended nor used as a "trick" for 
the purpose to deceive the audience, but rather to make possible the 
photographing of scenes which, if filmed by conventional methods, 
would prove too difficult, too dangerous, or too costly to be con- 
sidered practical. As such, the transparency process is not a 
miraculous "cure-all," but, intelligently and properly used, it can 
and does add immeasurably to the scope and efficiency of motion 
picture production. 

Regarding the future general use of triple-projection equipment, 
while its operation requires longer time and a more carefully set up 
organization, the resultant increased quality and scope amply justify 
its use. It is the author's opinion that the system makes a definite 
step forward. 

There will always be the problem of how to accomplish the many 
difficult and apparently impossible things that are written into scripts. 
However, our present-day process technicians are outstanding in their 
resourcefulness and versatility. They have the potential ability and 
capacity to be one of the industry's greatest economic assets of the 

184 A. F. EDOUART 

future. Their aptitude and skill in combining the real with the fic- 
titious, maintaining precise and critical realism, are amazing, and are 
limited only by the capacity of their equipment to perfectly accom- 
plish their ideas. Their understanding and appreciation of the dra- 
matic, artistic, and economic values in combining a photographic 
process embodying mechanical, illuminating, and chemical engineer- 
ing, place the transparency and special process branches of the 
cinematographic profession in an unparalleled position. 

With the advent of new and improved equipment, the physical 
scope and photographic effectiveness of the transparency process 
will definitely make further notable advances. With these advances, 
as with the present status of the process, the key to success, as 
already stated, is and always will be the technical expertness and 
artistic skill of the individuals utilizing and operating it. Such 
adroit performances have, in a very few years, brought the trans- 
parencies to their present excellence; while with improved equip- 
ment, allowing greater scope for the artistic imagination and better 
results technically and mechanically, the process seems bound to 
become a yet more useful instrument for the improvement of modern 


D. B. CLARK** 

Summary. Consistency in negative printing values is one of the most desirable 
factors in modern cinematography. Photoelectric light-measuring devices help the 
cinematographer maintain such consistency to a far greater degree than is possible 

Several requirements in these devices must be recognized, among which are freedom 
from error and photocell fatigue, changes in humidity or temperature, and the 
like, and good uniformity. 

While these requirements are not wholly met in existing meters, it has been found 
possible to use such meters to advantage. Coordination is effected by use of a special, 
portable testing unit of the photometer type. Further developments should include 
complete acceptance of strict time-and-temperature methods of negative development 
and some form of automatic, photoelectric-cell-contr oiled print-timing. This would 
remove all variables from the processing problem, and leave the responsibility for 
results solely in the hands of the cinematographer. 

Probably the most desirable single quality in modern studio cine- 
matography is consistency in negative printing values ; in other words, 
the ability to maintain throughout all the scenes of a production a 
uniformly correct exposure level, so that no corrective modifications 
in negative development or printing need be made. 

Such consistency would be desirable from almost every possible 
viewpoint. From the purely theoretical viewpoint of the engineer, 
it has long been axiomatic that a completely accurate reproduction of 
a scene is possible only when every factor from negative exposure to 
the printing of the positive maintains a normal relationship to every 
other factor in the chain, and that alteration of any of these factors 
may be offset, but can not be completely corrected, by modification of 
the other factors involved. 

From the viewpoint of the practical cameraman, maintaining 
this sequence of normal relations unbroken would mean positive 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
16, 1939. 

** Twentieth Century-Fox Studio, Hollywood, Calif. 


186 D. B. CLARK [j. s. M. P. E. 

assurance that the photographic effects obtained on the set would be 
preserved in the positive print. In other words, the cameraman 
would be, in truth, master of the results obtained on the screen, free 
from the fear that misunderstandings in the studio laboratory, or 
carelessness in the release-print laboratory might, as has so often 
been the case, distort his photography. 

From the viewpoint of the studio or laboratory operations execu- 
tive, labor, time, and money could be saved by reducing laboratory 
operations to a simplified norm. 

American cinematographers have deservedly been acclaimed for 
their remarkable consistency in this respect. But this consistency, 
great as it is, is relative, rather than absolute, due to the inevitable 
element of human fallibility. Visual fatigue and any of several other 
factors introduce minor variations from day to day and from one 
set-up to the next. These variations are well within the laboratory's 
corrective range, but nevertheless they represent a departure from 
the ideal. 

For this reason modern photoelectric light-measuring devices can 
be regarded as one of the most potentially valuable developments of 
recent years, substituting, as they do, an untiringly accurate electric 
eye for the customary visual judgment. 

However, to make practical use of such devices, it is not enough 
that the device merely exist. Methods of applying the device to 
the practical problems of the cinematographer must be worked out, 
and methods of coordinating the meters used in a studio with that 
studio's processing methods and with each other must be evolved. 
This we believe we have accomplished, so far as is possible with 
existing commercial meters, at the 20th Century-Fox Studio. 

Before describing these methods, however, it may be well to clarify 
the relationship existing between photoelectric light-measuring 
instruments and studio cinematographic methods. While for con- 
venience these devices are generally referred to as "exposure meters," 
such is by no means their purpose when applied to studio cinematog- 
raphy. On interior scenes especially, exposure is a constant fac- 
tor, while illumination is varied to obtain the desired exposure level. 
In other words, instead of following the practice general elsewhere, 
and modifying the lens aperture to suit the illumination available, the 
studio cinematographer keeps his lens opening almost invariably con- 
stant, and modifies his illumination to suit the lens aperture chosen. 
A basic reason for this is that the scenes must be intercut, and 


the varying optical quality given by constantly varied apertures would 
in most cases tend to prevent the scenes from maintaining consistent 
visual quality. 

In the same way, the professional cinematographer finds little or 
no gain from a light-measuring instrument that gives him merely an 
overall reading of the average illumination of a scene. Many papers 
presented at previous Conventions of this Society have stressed the 
fact that no two cinematographers employ the same lighting levels 
or the same lighting technic. The technical and artistic stock-in- 
trade of the studio cinematographer is his method of balancing light- 
ing, exactly as the brush-work of a Rembrandt or a Corot is, so to 
speak, the painter's individual trade-mark. It is entirely possible 
that two cinematographers might on a given scene use the same 
overall average of illumination, yet because of their different methods 
of balancing the light obtain startlingly different results on the 

The basis of modern lighting technic is the so-called system of 
"key lighting." In this, the principal illumination of the scene, usu- 
ally that falling on the face of the most dramatically important player, 
is considered the key. All other gradations of lighting, both in 
stronger highlights and in shadows, on players and setting, are 
balanced to this key light. If this key light is correct, the balancing 
of the scene, and with it, the overall exposure value, will be correct. 
If it is incorrect, the entire scene can be thrown off balance. 

Therefore we at the 20th Century-Fox Studio, in common with 
the majority of meter-using cinematographers, have found it best to 
use the meter simply to measure this key light. This method has 
proved to be the most economical of time and effort. In addition, 
it leaves the cinematographer free to balance his lighting as he sees 
fit, thus completely preserving his artistic originality. Yet it gives 
him the benefit of a starting point that he knows to be accurate. 

But to gain these benefits, both the method of using the meter and 
the meter itself must be accurate, with all possible variables, both 
personal and mechanical, minimized or if possible eliminated. Under 
studio conditions, using a meter for reading the light reflected from 
the subject seems susceptible to considerable error. It is too easy 
for the meter-reading to be made inaccurate by the direct rays of 
some one of the many supplementary lighting units behind, above, 
or to one side of the subject reaching the cell of the meter. In 
some instances, too, the meter may accidentally be reading in its 

188 D. B. CLARK [j. s. M. p. E. 

own shadow, again introducing an error. Further, the acceptance 
angle of a reflection type meter is seldom sharply denned, and when 
such a meter is used under production conditions, it can introduce 
considerable error. Therefore, we have standardized on direct- 
reading meters. 

But this is only the first step. Such meters must be inherently 
accurate, and some means must be provided for checking that ac- 
curacy at frequent intervals, and for coordinating the meters with the 
methods of the laboratory processing the film. 

Before standardizing on any type of meter, we made tests of a 
number of samples of meters of various makes over a considerable 
period. The tests were made on a simple optical bench, in which the 
meters were always tested at a measured separation from a light- 
source operated at constant intensity. The factors of photocell 
fatigue, effects of changing weather-conditions (including humidity) 
and the like, were carefully noted. Variations between individual 
meters of the same manufacture were particularly observed. 

While it would be far too optimistic to infer that the ultimate in 
meter consistency has yet been attained, a commercial type suffi- 
ciently accurate for our use was found. While still subject to error 
especially in cases of changes in humidity the type selected proved 
the most nearly consistent of all those tested, and accordingly has 
been accepted as the studio's standard equipment. Meters have 
been provided by the studio for the use of all the staff Directors of 
Photography (First Cameramen). This is believed to be the first 
instance of a studio's providing such instruments for its staff. 

It should be emphasized here that while the meters have been 
provided for all the staff, no attempt is made or will be made to force 
their use. This decision is left strictly to each individual. The 
record of those who have accepted the meters and used them cor- 
rectly, however, is such that virtually all the staff have of their own 
volition accepted the aid of these instruments. 

While the optical-bench testing set-up used in these initial tests 
was technically accurate, it lacked the simplicity and portability 
desirable for every-day use. Therefore Grover Laube, of the Studio's 
precision machine shop, and the writer developed a convenient test- 
box which, if necessary, can be easily carried to any set or location 
where an immediate check of a meter is needed. 

This device consists of a small wooden case which houses a bat- 
tery-operated automobile headlight bulb, which is used as a light- 


source. Between the light-source and its power supply are inter- 
posed an ammeter and a rheostat. The bulb is fixed rigidly in 
one position, and has a long useful life; and by applying a known 
current to it, as indicated by the ammeter, its light-flux will be con- 

At one end of the test-box is an aperture, faced with a suitable 
ground-glass diffuser, to which the meter being tested is applied. A 
suitable shield fits closely around the meter-casing and excludes all 
external light. 

In testing a meter, the instrument is applied to the aperture of the 
test-box and two readings are taken: first a zero reading, with the 
testing light off; second, the testing light is turned on and brought to 
its predetermined normal intensity by manipulating the rheostat. 
At this intensity the meter being tested should give a predetermined 

A wider range of test readings would easily be possible; but in 
view of the methods of using the meter in the studio, such does not 
at present seem necessary. 

As has been explained, we have found it desirable to measure only 
the key light, thereafter leaving the cinematographer free to balance 
the rest of his lighting visually, as he may see fit. Superintendent 
Leshing, in charge of the studio's laboratory, has found as every 
laboratory man has that regardless of lighting balance or of the 
effects sought, under the conditions applying at any given laboratory, 
the most faithful result will be obtained by having a negative that 
will print on a given normal printer-light setting. Precisely which of 
the twenty-two printer lights is available for this normal may vary 
considerably among different laboratories, as it is affected by the 
processing methods and equipment used at the plants, including such 
variables as negative developing solutions and methods, positive 
development, and in some cases not only the types of printers used 
but also the modifications made in these printers to coordinate them 
with the plant's standards. Since any concrete figure I might give 
would apply only to the 20th Century-Fox Studio's laboratory, 
suffice it to say that we prefer a negative that will print very closely 
in the middle of the printing scale. 

Thus if our cinematographers know that their key light is pinned 
rigidly to a value that will allow negative printing normally on this 
most favorable printer-light, they can balance the rest of the lighting 
to suit the effect they want, confident that both negative and posi- 

190 D. B. CLARK [J. s. M. P. E. 

tive will receive strictly normal and favorable treatment in proc- 
essing, and that the result on the screen will be as desired. 

Therefore all that is necessary is to determine a point on the 
meter's primary scale that will, when the meter is used for direct 
reading on the key light, correspond to an exposure-level that will 
give this optimum printing density. As has been already pointed 
out, while it would be perfectly possible to take additional readings 
of the shadows, background, and so on, it is not necessary due to the 
skill with which a capable studio cinematographer can balance light- 
ing visually. On the other hand, the meter's guidance in assuring a 
constantly normal starting point in the key light is invaluable. 

The sunshade fitted to the type of meters we use has proved a 
valuable feature under studio conditions, as it excludes unwanted 
rays from lamps other than those producing the key light. The 
slotted cover of this hood, intended by the manufacturer for reduc- 
ing high intensities to fractions that, while still accurately measur- 
able, will not overload the photocell, is not so useful to us, as it makes 
the meter too strongly directional for our purpose. Therefore we fit 
a metal plate having a small rectangular aperture directly over the 
meter's inner opening. This cuts down the illumination activating 
the cell in the same proportion, but at the same time does not ex- 
aggerate the directional characteristics. 

In practice, the meter is used in a position close beside the face of 
the subject, and a direct reading taken on the key light. If this is 
found to have the correct value for normal printing, all is well. If 
not, the key light is raised or lowered to give th desired reading, and 
the balance of the lighting modified accordingly by the usual visual 

As a general rule, the meter-reading is most frequently taken only 
when the lighting set-up is virtually completed, since experience 
enables most cinematographers to read lighting visually with great 
accuracy, and the meter need be used only as a check. 

It may well be asked, "Does this use of photoelectric meters im- 
prove the consistency of a studio's cinematography?" The answer 
in our case is a definite affirmative. 

Before the system was adopted, an extensive series of tests was 
made. On two successive days, a cameraman was sent to film identi- 
cal scenes. His only instructions were to make long-shots, medium- 
shots, and close-ups, always keeping the meter-reading of the key 
light on the predetermined figure. The third day he was sent to make 


similar tests on a different set, of different coloring and design, with 
different actors and different costumes. The fourth day this test 
was repeated. The fifth day he was assigned to make tests of ex- 
terior night-effect scenes. To summarize briefly, every take printed 
on the desired printer-light setting, and showed uniformly satisfac- 
tory negative values. 

Since then the system has been used by an increasing number of 
the studio's camera staff. Where previously virtually the entire 
range of printer-light adjustments had to be used daily in making 
rush prints of the various scenes made by the various men, today a 
maximum range of three to four printer-lights will accommodate 
virtually all the variations encountered on normal interior scenes. 

Here, for example, is the laboratory's printing record for the last 
nine productions completed at the studio, expressed in terms of over- 
all printing averages for all the footage printed as daily prints: 
The first three printed on an average of 12+ ; the next three averaged 
on light 14; the next two, both of which included more than the 
ordinary number of special effect-lightings, averaged on light 15; 
the ninth, a melodrama with an uncommon proportion of abnormal 
effect-lightings and night exteriors, averaged on light 17. The 
average printer-light setting for the nine productions was 13.8. In 
other words, these nine productions could virtually have been printed 
on one printer setting say, light 14 without serious harm. 

This average must be qualified yet further. The averages quoted 
above are those obtained in making the daily or rush prints the 
first prints made of the scenes, for use in cutting and in checking the 
acceptibility of action, etc. Such prints correspond roughly to a 
portrait photographer's proofs; they are not the finished product. 
Almost inevitably the timing of at least some of the rush-print scenes 
proves to be inaccurate, especially in the case of related scenes or 
sequences made some days or weeks apart, and printed without 
specific consideration of the contiguous scenes the cinematographer is 
striving to match. When the master print and the release prints are 
made, these inaccuracies are equalized, and it is certain that the aver- 
ages for the release prints of the same nine productions would fall 
within even closer limits. 

The assistance of the meters can be given much credit for the ease 
with which our cinematographers have accomplished the recent 
transition from yesterday's slower films to today's ultra-fast emul- 
sions. Before the film was put into general use, Mr. Leshing and the 

192 D. B. CLARK [j. s. M. p. E. 

writer made tests to determine the best photographic and laboratory 
treatment for the new emulsions. In the course of these tests, the 
meter-readings corresponding to the best photographic values were 
determined. Thus when the men were given the new film for produc- 
tion use, they were also furnished these data. With this positive 
information as to the normal key light requirements of the new film, 
their knowledge of lighting balances and their trained visual judg- 
ment enabled them to make the transition with perfect assurance and 

The same general technic of using the meter can be applied with 
equal benefit to the making of exterior scenes. However, some modi- 
fications are necessary due to the changed conditions. No practical 
meter commercially available in this country can be used for taking 
a direct reading on the sun, which is of course the actual key light 
on an exterior scene. Further, while the cinematographer can ex- 
ercise some degree of control over exterior lighting, he can do so only 
over a relatively restricted area, usually in the foreground of his 
shot. The background can not be controlled to any great extent. 

For this reason, in making exterior scenes we consider this virtu- 
ally uncontrollable background illumination as the key light, and 
take a reflection reading on this, positioning the meter to exclude the 
immediate foreground. Using this as the known and relatively fixed 
factor in our problem, we can manipulate the foreground lighting 
especially that upon the players to give us balanced exposure values. 

At the 20th Century- Fox Studio we are singularly fortunate in our 
attempts to apply this system. It is a real asset that the studio's 
laboratory adheres to a strict time-and-temperature system of nega- 
tive development. With such a system we know that, granted con- 
sistently correct exposure values, which we are obtaining with the 
meters, and consistent negative development, gained from the time- 
and-temperature methods followed, we should be able to count on 
consistent printing values for our negatives. 

It is to be admitted that there still remain some variables that all 
of us would like to see overcome. There will, for instance, probably 
always be some slight variation in the way individual cameramen will 
read their meters a few inches' difference in the position at which the 
reading is taken can mean a difference of two or three printer-light 
values in the resulting reading. There are also minor but cumulative 
variations possible in negative processing, as, for instance, in solution 
strength, temperature, etc., which despite all reasonable precautions 


will vary slightly from day to day. But the worst variations have 
been taken out of photographing and negative processing. 

The same is true of the printing and development of the positive 
film. The greatest remaining variable is in the "timing" or deter- 
mination of printer-light settings for positive print-making. This is 
done visually, and is subject to lively variations which could very 
advantageously be eliminated. 

The first of these is the element of visual fatigue on the part of the 
print timer, so familiar it need hardly be dwelt upon here ; second is 
visual misjudgment. Visual print-timing is almost always based 
upon estimation of the printing value of face- textures of the prin- 
cipal players. This is subject to considerable error at times, for 
contrast between face areas and adjacent densities may be highly de- 
ceptive, even to the trained eye. 

A very simple illustration will prove this. Consider three pairs of 
concentric circles; let the center circle in each set be of an identical 
pure white. Let the outer circles be respectively black, light gray, 
and dark gray. It will be found that in every case the white circle 
surrounded by the darkest black ring will look whitest, while the one 
surrounded by a ring of light gray will seem a dirtier white. This 
same optical illusion affects the visual judging of small areas, such as 
faces, in timing motion picture prints. 

Third, and often the most irritating, is the element of misunder- 
standing between the laboratory and the cameraman as to what is 
sought in the scene. Lacking specific instructions to the contrary, 
the timer may jump to the conclusion that the cameraman is seeking 
an effect entirely different from his real aim. To cite an extreme 
instance, night-effects have sometimes been known to be printed up as 
day-effect shots. More common is the less obvious error of printing 
a scene too light or too dark. This often passes unnoticed in its 
true aspect, and results instead in an impression that the man at the 
camera was at fault in lighting or exposure. 

Since the general acceptance of meters by our studio's cinema tog- 
raphers, the writer has frequently noticed such instances of mis- 
interpretation. Viewing the "dailies," the comment would be that 
such-and-such a scene or sequence was too dark or too light. In 
almost every instance since the use of meters has been general, in- 
vestigation has shown that the camera crew, guided by the meter, 
lighted the scene for normal printing; but misjudgment in print- 
timing had caused the scene to be printed incorrectly. A reprint, 

194 D. B. CLARK [J. s. M. P. E. 

made at the normal printer setting, has almost invariably shown that 
the cameraman, backed by his meter, was right, and the visual judg- 
ment of the timer was wrong. 

Is it not logical to expect, therefore, that ultimately some satis- 
factory method of applying the photoelectric cell to print-timing can 
be introduced? Certain types of 16-mm reversal film are already 
processed with an automatic control of this type which adjusts the 
flashing or printing exposure to the overall transmission value of the 
developed negative image. It is possible that this precise method 
might not prove satisfactory for professional use, due to inability to 
make the allowances necessary for special light-effect scenes. But 
could it not be possible at least to develop a technic of reading nega- 
tive face-values by means of a semi-automatic photoelectric densi- 
tometer that would eliminate the element of human fallibility, and give 
a reading, not in densities, but directly in terms of printer-light set- 

The practical results of such a development, when coupled with 
consistent time-and-temperature negative processing and the con- 
sistent photographic results obtainable with the use of coordinated 
meters, should be of worth-while practical value. Laboratory opera- 
tions would be simplified, and greater consistency and economy ob- 
tained. The cinema tographer would have an absolute normal to 
which to peg his lighting. Granted such a standard, free from varia- 
tion, he would be free to exercise his creative individuality without 
fear that subsequent variations or misunderstandings in the labora- 
tory would possibly nullify his efforts. 

Admittedly this would place the responsibility for results exclu- 
sively on the shoulders of the cinematographer. That, however, is 
where it belongs and where he wants it. The guidance of the 
meter would keep him within the mechanical tolerances set by the 
limitations of emulsion and processing. The rest would be solely 
up to the man behind the camera, to his judgment of lighting bal- 
ance, and to his artistic skill. 

At present, we have made commendable progress toward this 
goal with the increasing acceptance of meters, and, I believe, with the 
development of the methods here outlined of testing and coordinat- 
ing meters at the 20th Century-Fox Studio. It is a demonstrable 
fact that the confidence of our cinematographers and the consistency 
of the results they obtain has been improved by adding to their 
acknowledged skill the guidance and mechanical accuracy of the meter. 


In closing, I would like again to stress the fact that while today's 
meters are good, they do not by any means represent perfection in 
meters for studio use. In making this statement I do not overlook 
the fact that the two types of photocell meters almost exclusively 
used by monochrome cameramen were designed primarily for ama- 
teur use, and that other models, designed more primarily for labora- 
tory use, offer certain technical refinements impossible in these 
smaller, lower-priced meters. However, there are certain require- 
ments that a meter for use under modern conditions in a motion pic- 
ture studio should fulfill; some of them, as will be seen, rule out most 
of these advanced laboratory- type instruments. These require- 
ments include : 

(7) Dependability. The ability to withstand at least moderately rough usage. 

(2} Uniformity. Consistent accuracy, sufficient so that all the meters used 
by a studio can be expected to give, under comparable conditions, readings 
sufficiently comparable as to remain within a minimum corrective range in the 
printing process. 

(3) Freedom from Variations Caused by Change in Humidity, Temperature, 
Etc. This at present is not always the case. Tests of our own meters showed 
that all of them read several printer-lights below normal under conditions of 
low humidity, and above normal when humidity is high. In my own previous 
experience, using a meter on location in the South Seas, where temperature and 
humidity were both high, the meter read abnormally high. Members of the 
Byrd Antarctic Expeditions have informed me that the many meters taken to 
Little America by both professional and amateur photographers in the party all 
gave uselessly low readings in the antarctic. 

(4) Wider Sensitivity. This is especially needful in the case of meters used 
in connection with projected background cinematography. In this process, the 
illumination transmitted through the background screen is necessarily the factor 
to which all foreground lighting must be keyed. The meter-reading upon this 
must inevitably be taken with the photocell directed upon the screen, from the 
same side as the camera. In the tests we have so far made we have not as yet 
succeeded in finding a meter capable of giving us such a reading, despite the fact 
that at the same time the foreground camera was indisputably receiving sufficient 
light through the screen to make a satisfactory exposure. The use of meters on 
these scenes would be of especial value. 

(5) Compactness. For studio use, a meter must be small, convenient, and 
inconspicuous, for several reasons. A small, handy meter that, like the ones 
now in use, can be carried easily in the pocket, will be used much more regularly 
than any larger device. In addition, the disturbance necessarily incident to 
bringing a large meter into the set and using it tends to upset even the most even- 
tempered of players when trying to concentrate on action and dialog. It must 
be confessed, too, that some directors men of sufficient reputation and experi- 
ence to know better have been known to grow sarcastic when their cameramen 
employ such aids, though they themselves take full advantage of every possible 

196 D. B. CLARK [J. s. M. P. E. 

aid to their own work, such as using public address systems whenever there are 
more than four or five players in a scene. Under such circumstances, the small, 
inconspicuous meter is the only one likely to be used. 

Another important improvement would be, in the case of reflection-type meters, 
an instrument with an acceptance angle more closely coordinated with that of the 
camera. Many such meters have an angle as great as 60 degrees, while the two 
most commonly used objectives, the 40-mm and the 50-mm, have horizontal 
angles of 30 and 25 degrees, respectively. Some form of finder would be a de- 
sirable addition, and one which should be even more useful in the wider amateur 
field, as well. 

In general, however, it must emphatically be stated that during the past year, 
especially, the industry has made great progress in adapting the electric eye of 
the meter to studio use. It has been most spectacular in the way it has facili- 
tated the change from conventional to fast films. It has increased the con- 
sistency of every cameraman's work, and freed him from the ever-present fear 
that physical or visual fatigue might be distorting his judgment of lighting. 
It is relieving the First Cinematographer of the burden of routine work and 
leaving him more free to exert his creative artistry. It is therefore logical to 
expect that as meters continue to gain in acceptance, individuals become more 
accustomed to their use, and as improved and more dependable instruments are 
developed, we should witness magnificent advances in both the art and the 
science of cinematography. 


DR. MILLER: The subject of accurate control of exposure and processing of 
motion picture negative is one that has received a great deal of attention within 
recent years. Accurate processing control has been demanded since the intro- 
duction of sound recording on film, and the general adoption of sensitometric con- 
trol of processing has served greatly to improve both picture quality and uni- 
formity. It has only been within the last few years, however, that the cinema- 
tographer has been afforded an instrumental means of determining proper ex- 
posure, and it seems inevitable that here again the substitution of instrumental 
control for human judgment will result in further improvement in picture quality. 

Mr. Clark, what degree of variation in sensitivity has been noted among the 
various meters employed? Is the percentage variation in sensitivity of rather 
high value or are the meters reasonably uniform? 

MR. CLARK: The meters produced by a given company, for instance, may vary 
as much as three or four printer points in reading and when tested alongside each 
other. On exteriors I have known five meters to vary as much as one hundred per 
cent in exposure. 

MR. SKINNER: I use incident light on both interiors and exteriors. The same 
factor can then be followed all the way through, as the meters are sensitive enough. 
It is a matter of reducing their sensitivity so they can be used for exterior light as 

MR. CLARK: Measuring reflected light on exteriors is very easy, so we have 
never tried reducing the sensitivity for measuring direct sunlight. 

MR. HYNDMAN: Present available photoelectric meters for measuring light- 
intensity have photocell reception cone angles of approximately 50 to 60 degrees, 


and consequently it is practically impossible, except at very close range, to make 
a light-intensity measurement of the reflected light from a definite small area of 
the subject. In other words, to measure the light reflected from a finite, small 
area necessitates placing the meter within a few inches of the subject, otherwise a 
large area is included in the measurement. On a studio set it is possible to control 
the lighting by photoelectric measurement of the intensities reflected from various 
areas of the subject, whereas it is often difficult to control daylight or sunlight 
falling on a subject or series of subjects out of doors. The problem of the proper 
use of photoelectric meters on studio sets and out of doors is quite different, but 
with intelligent application present meters can be used satisfactorily in both cir- 
cumstances. To measure a small finite area on a given subject with a photoelec- 
tric meter at a distance of several feet would necessitate using a meter with a cone 
angle entrance pupil to the cell of approximately 3 to 5 degrees, but unfortunately 
available photoelectric cells do not have sufficient sensitivity to make reflected- 
light measurements in these circumstances with normal lighting conditions. It 
is therefore impossible with the present type of photoelectric meter to measure a 
small area on a given subject from the position of the camera which may be 
several feet from the object of interest. Furthermore, it appears that a meter 
that would accomplish this purpose will not be available until the sensitivity of the 
photocell is materially increased. 

MR. PALMER: With regard to measuring incident light in color photography, 
the cells commonly used in photoelectric meters are not equally sensitive to all 
colors. It is possible to get rather erroneous readings on subjects of different color. 
When using a meter such as Mr. Clark suggested in connection with black-and- 
white photography, we rule out the question of non-uniform sensitivity of the cell 
to color. My experience in taking readings for color exposures has shown that the 
results are more reliable by incident light than by light reflected from the subject. 

MR. SKINNER: It appears that Mr. Clark described two different methods in 
his paper; has one any particular advantage over the other? 

MR. CLARK : We have had satisfactory results on exteriors only by measuring 
reflected light. We do not get satisfactory results with reflected light on interiors 
although we have instruments that will read reflected light. The meters are not 



Summary. Practically all precision sound-track measurements are made with 
a microscope, which has been the accepted standard in the industry. This method 
is necessarily slow, and in an effort to speed up track position, printer alignment, 
and other technical measurements, a projection microscope has been developed. Using 
equipment available at a reasonable cost, a projection microscope which can be used 
by anyone in the sound and laboratory departments has been developed and is herein 

One of the most important advantages of the instrument is its ability to detect print- 
ing machine defects, and by its use the quality of sound printing has been greatly 

Since sound was first recorded on film, the instrument by which pre- 
cision adjustments of recording machines, printers, and sound re- 
producers are made has been the microscope. This instrument will 
be found in most of the sound department engineering staff offices 
and in the sound-track control rooms of our laboratories. 

The microscope generally used is 100 power, with a 50-power ob- 
jective additionally provided for non-precision work. By means of a 
very accurate jig, and a scale calibrated in 0.001-inch divisions in the 
microscope eyepiece, the negative or print to be measured is mounted 
in the jig and moved back and forth until the section of track that 
is to be inspected is within the field of vision. With the 100-power 
objective, the actual field of vision is a circle about 50 mils in di- 
ameter, so that the entire sound-track can not be viewed at one time. 
The 50-power objective will permit viewing the entire track of 100 
mils, but the scale becomes so difficult to read as to preclude accurate 

To overcome this handicap and provide an instrument that would 
not require a trained eye to read the results, the writer began experi- 
menting with various types of optical projectors in an effort to find a 
combination that would replace the microscope hi this type of work. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 18, 1939. 

** Warner Bros. -First National Studios, Burbank, Calif. 



We are all familiar with the method frequently used of projecting 
sound-track in a standard picture projector, to watch overloads or 
printer weave ; but this method has the handicap that the film passes 
intermittently through the aperture at the rate of 24 frames per 
second and can not be stopped for examination and measurement of 
a small section. 

However, as the standard pressure-plate, shoes, and guides of a 
picture projector seemed to be the best method of supporting the film, 
it was decided to use this method of film suspension as a nucleus 
around a suitable optical system and source of light. The finished 
projection microscope is shown in the accompanying illustration. 

The source of light was obtained by modifying a Spencer Type C 
Delineascope, a projector intended to show single frames of motion 
picture film on a screen about 2X3 feet in size. This equipment 
consists of a 500-watt lamp in a compact housing, with condensing 
lenses of special heat-reducing glass so as to concentrate sufficient 
light while removing most of the heat from it without water-cooling 
or air-blast. The film-slide attachment and optical system of the 
Delineascope is not suitable for sound-track, so it was not used. In 
its place, the gate and aperture plate of a Simplex projector were 
mounted on a heavy base, the aperture being masked down to a 
slot 0.20 inch wide and 4 sprocket-holes long, with a slot cut out to 
show the extreme edge of the film. Th lens used is a lVf-inch//1.5 
objective, which, with a throw of 14V2 feet, will produce an image 
equivalent to one frame of sound-track magnified 120 times. 

By setting the aperture plate so that the sound-track center is in 
the center of the light-beam, sufficient illumination is obtained to 
permit very accurate measurements of sound-track position, if the 
room is well darkened. Naturally a superbrilliant image can not 
be projected to such large dimensions with a 500-watt source, but 
this amount of light is the maximum permissible without fire hazard. 
Additional protection against fire was provided by placing a quiet, 
low-speed ventilating fan in an extended stack on top of the lamp 
house, and by building a protecting metal screen around the lamp 
house to keep the film from touching the hot metal parts. With this 
set-up, film can remain motionless for 10 minutes in the light-beam 
without overheating or curling. 

A pair of rewinds are mounted above and below the projector, which 
is, in turn, mounted on a platform bracketed to the wall, thus per- 
mitting a long reel of film to be run and stopped at various points 



LT. S. M. P. E. 

wherever inspection is required. The> concave mirror provided in the 
lamp house produced a double image when high-frequency recordings 
were projected, so that it was necessary to front-silver the mirror 
to eliminate this effect. 

For a screen, the wall on the opposite side of the room was coated 
with a flat, white paint, and a scale permanently painted in the cen- 
ter of the screen, as shown in Fig. 1. This scale marks the center-line 
of the track, the position of the bias lines in bilateral variable-width 
recording, the limits of 100 per cent modulation, and, at the edge of 


L s 




FIG. 1. 

Scale for checking sound-track with 
projection microscope. 

the screen (not shown in Fig. 1) is a base-line representing the guided 
edge of the film. The objective lens is mounted on a micrometer base, 
with an adjusting worm to move the lens back and forth until the 
guided edge of the film coincides accurately with the base-line on the 
screen. Using the scale mentioned above, track position, bias width, 
percentage modulation, and all other dimensions can be accurately 
measured. In addition to the horizontal scale, a vertical scale of 
frequency calibration is included, this scale permitting the identifica- 
tion of the speech or music frequencies that may be causing trouble. 
Often an overload is caused by a single frequency at periodic intervals, 


and by this scale the frequency can be determined within a reasonable 
range by setting the tip of one of the striations on the base-line X and 
observing where the next striation tip falls on the scale. 

Another use for the projector is in measuring the opening and 
closing times of noise-reduction shutters, negatives of shutter opening 
and closing tests being projected on a special scale temporarily super- 
imposed over the regular scale. 

One of the principal advantages of a projection microscope is that a 
number of persons can view the sound-track at the same time. Fre- 

FIG. 2. Projection microscope. 

quent use has been made of the projector to demonstrate to groups 
during technical discussions, and mixers, engineers, recording ma- 
chine maintenance men, and laboratory supervisors are expected to 
use the projector whenever needed. But by far the greatest use to 
which the projector is applied is checking the printing machines in 
the laboratory. Quite by accident it was discovered that if a 9000- 
cycle frequency print were moved past the aperture at a rate of about 
2 frames per second, any printer slippage or lack of contact would 
show up on the screen in the form of "ghosts" or dark patches. A 
printer that has perfect contact and no slippage will produce a 9000- 
cycle print which when run at 2 frames per second on the projection 
microscope will produce a uniform blurred image that does not vary. 
A poor print will flicker in much the same manner as a picture pro- 

202 G. M. BEST 

jector with a light-source of alternating-current carbon arc, with the 
shutter not properly tuned. 

It is now the custom at Warner Bros. Studios to project daily a 
sample 9000-cycle print from every printer to be used during the day, 
thus reducing printer troubles to a minimum due to the better super- 
vision obtained. The RKO parallel-line test, which consists of a 
series of 1-mil exposures approximately 2 mils apart across the full 
width of the track, is also useful when used in conjunction with the 
9000-cycle contact test, the same ghosts appearing in this track as 
are observed on a frequency print. Too rapid motion of the film eli- 
minates this effect, so that a little practice will soon tell the operator 
at what speed to propel the film to get the best results. Printer weave, 
the presence of 96-cycle modulation due to poor contact around the 
sprocket-holes, and poor track illumination in the printer are all easily 
detected, and several of these projectors have been built by foreign 
laboratory heads who have seen it, with notable improvement in their 
product after being put to use. 


DR. FRAVNE: Can you tell us anything about this printer? 

MR. BEST: The printer used for all the tests shown in this demonstration 
utilizes the RCA non-slip principle, and there is nothing about it that has not 
been adequately covered in papers previously presented to the Society. The 
demonstrations of bad printer slippage were made on an old sprocket-type printer 
which has been in service for seventeen years and is of a type no longer generally 
used for high-quality sound printing. The sample of 96-cycle noise caused by 
poor printing was produced by disturbing the adjustment of the drum roller 
so as to cause contraction of the film around the sprocket-holes. Improper ad- 
justment of this roller produces 96-cycle difficulties found hi practice, and it 
is very easy to put the roller out of adjustment deliberately. 

MR. KELLOGG : Were you able to find any way of changing the 96-cycle modula- 
tion? Do you think the burrs on the sprocket-holes actually push the film apart, 
or does it seem that the film bends into a polygon rather than a circular arc? 

MR. BEST: I think that is due entirely to the bending at the sprocket-holes 
due to weakening one edge of the film. The entire film from the dividing line 
between the sound-track and the picture to the edge of the film is not supported 
by a roller; it is free, and contact between the negative and positive depends en- 
tirely upon the excellence of contact through the picture area. By proper design 
of the drum, contact roller, and guides, and accurate adjustments of all three 
elements, the difficulty from 96 cycles can be reduced to such a small amount 
that it can not be detected. If the contact is poor at the extreme edge of the film, 
the effect may be apparent for ten or fifteen mils into the sound-track, in which 
case the level of the 96-cycle note may be so far down with respect to the signal 
that it is not audible in normal theater reproduction. 





Summary. Direct recording is becoming commercially more and more important. 
Acetate blanks are used for high-quality recordings, but these materials are essentially 
softer than pressed records, and therefore make necessary new considerations in the 
design of a high-quality pick-up to be used with them. 

It is shown that a dynamic stylus pressure of approximately 25 grams is the maxi- 
mum force that acetate can tolerate without permanent deformation of the modulated 
grooves, even when due consideration is given to the proper matching of stiffness and 
inertia of the vibratory system of the pick-up. 

A simple formula is given for the most suitable condition of the matching of inertia 
and stiffness for a complex wave-form. Other factors that interfere with the construc- 
tion of a light pick-up, such as uneven record and turntable surfaces, are explained, 
and suggestions are made for the reduction of these effects. 

The advantages of ''constant amplitude" as a method of recording and reproduction 
are shown, and a constant amplitude system is demonstrated. 

The phonograph is the earliest development in the art of sound re- 
cording. Since Edison built the first machine and visualized its great 
commercial value, the method of mechanical recording has been 
steadily expanding. As early as the last century, the phonograph 
has commanded public recognition. We find an interesting quotation 
from Scientific American of fifty years ago, which refers to this sub- 

"The improvements in the phonograph have now been carried to such a degree 
of perfection that the instrument is practically ready for general introduction. 
Undoubtedly means will be hit upon from time to time to enhance the value and 
efficiency of the phonograph, but it stands today, in our opinion, far more practical 
and complete than was the typewriter when first brought out and placed on the 
market. Back of all the tall talk and exaggeration on the subject. . . is a machine 
of admirable performance, whose utility is so wide and various that it is hard to 
determine just which work will give it the largest field of employment. . . . And 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received May 8, 

** The Brush Development Co., Cleveland, Ohio. 


204 A. L. WILLIAMS [j. s. M. P. E. 

then, too, it is the wonder. . . that not only can the human voice be registered, but 
it can be duplicated in countless electrotypes." 

There is scarcely another art that has undergone such extensive 
improvement as the technic of sound transmission and sound repro- 
duction. Quality acceptable a few years ago is not acceptable today, 
and what was considered a good sound recording a decade ago, would 
not be tolerated today. Without any doubt, the reason for this is 
that the sound picture and the radio have educated the public to de- 
mand better quality in recorded sound. This should not be surpris- 
ing, if we consider that every tenth person in the world is provided 
with a radio. Engineers and physicists have steadily put forth their 
efforts to investigate and analyze the requirements for good sound 
reproduction, and to find new ways to apply their knowledge to ob- 
tain further improvement. The method of mechanical recording so 
far, however, has been commercially exploited only in the manufac- 
ture of disk records. 

It is somewhat surprising that the use of direct recording, with the 
exception of dictating machines, has not found as much acceptance as 
it should be entitled to. But nevertheless, quite a few companies 
nowadays are concentrating their efforts toward improving and pro- 
moting instruments of this kind. The writer is of the opinion that 
direct recording will eventually be as commercially important as the 
manufacture of completed disk records. 

In the light of these considerations, this Company has given a con- 
siderable amount of attention to perfecting electroacoustic apparatus 
for sound reproduction. In the past few years, experience has taught 
us that direct recording has not only considerable entertainment 
value, but also that it has important educational adaptations. The 
entertainment possibilities of such a vehicle are very obvious, since 
it means that favorite radio programs can be preserved, and the music 
enthusiast is enabled to build up a musical library of his own choice. 

The educational possibilities are being more widely recognized 
today, and this fact is reflected by the acceptance of direct recording 
machines by many schools and colleges, orchestra leaders, and artists. 
There is every reason to believe that the preservation of sound will in 
the near future be as important as the preservation of appearance. 
What the candid camera has done commercially for photography, 
the instantaneous sound recorder may well do for its industry. It 
seems timely, therefore, to analyze the problems involved in setting 
up a machine for high-quality instantaneous recording. 


Since in the art of mechanical sound recording the laterally cut 
record has a predominant place, this paper, for the sake of simplicity, 
will restrict itself to considerations of this method. Regardless of 
whether a disk, a cylinder, or a film is used, the problems are very 

The method of lateral recording requires that the sound carrier be 
in relative motion to the cutter or the pick-up. The record is made 
by cutting or embossing the signal to be recorded. The cutting or 
embossing stylus moves in the plane of the disk perpendicularly to the 
relative motion of disk material, and causes modulation of the groove 
corresponding to the vibrations of the signal. In the process of repro- 
duction, a stylus is guided by these modulated grooves, and being 
forcefully moved in accordance with the modulations, transfers them 
into electrical energy. 

The cutting stylus cuts a groove into the record in such a way that 
the groove walls are inclined to each other at an angle of approxi- 
mately 90 degrees. Such a groove has a V shape. Since it is impos- 
sible to provide an ideally sharp point on the cutting stylus, the stylus 
is rounded and has, for most commercial applications, a radius of 
0.0023 inch. If a disk is used as record material, the procedure is to 
cut in helical form. To provide a long-playing capacity for such re- 
cordings, it is desirable that the grooves be as close to each other as 
possible. While most commercially available records are cut with 
approximately 96 lines per inch, direct-recording machines are on the 
market that cut as close as 160 lines per inch. Since, as pointed out 
above, these grooves are laterally modulated by the recorded signals, 
the amplitude of recording is limited by the separation of the grooves. 
It is necessary to avoid "over-cutting," which may occur when adja- 
cent grooves are not sufficiently separated by the intervening wall. 
Even for the greatest signal amplitude, over-cutting must be elimi- 
nated. For this reason, the usual commercial practice is to cut "con- 
stant velocity" in the higher frequencies, and "constant amplitude" 
in the lower frequencies. The frequency at which the transition be- 
tween constant velocity and constant amplitude is made to occur is 
generally between 300 to 800 cycles. 

Instantaneous recording requires much softer record material than 
the so-called pressed records. More and more nitrate-coated disks 
are coming into use, and the probability is that even if new record 
materials are found, they will, mechanically, be very like the nitrate 
disks now available. It is easy to see that a different record material, 



[J. S. M. P. E. 

particularly since it is a softer one, miist necessarily call for a diff erent 
pick-up design. Frequent reproduction of a record can be satisfac- 
tory only if the forces that have to be supplied by the record to the 
pick-up are not great enough to deform the disk material perma- 
nently. It is necessary, therefore, to investigate these forces, with 
the view of determining optimum design features. But before this 
can be done successfully, the basic design of a pick-up will be briefly 
mentioned. A pick-up is an electric generator, whose mechanical 
force is supplied by the groove modulation through the stylus as- 
sembly to the generator element. Since vibratory motions are tak- 
ing place, the pick-up generator must respond to vibratory motion; 
consequently, each pick-up represents a vibratory system in itself, hav- 
ing a certain amount of stiffness and a certain amount of inertia. 

FIG. 1. Stylus resting in unmodulated groove. 

Fig. 1 is a diagram of a stylus resting in an unmodulated groove. 
It is under a pressure P, and both walls exert a force normal to their 
surface counteracting force P. In the vector diagram, the three act- 
ing forces are shown, and it is obvious that the forces exerted by both 
walls are equal. Referring to Fig. 1, 

P W b 

W a 

-- sn - =_ cos 

sin a \2 2/ sm a 2 

sin a = 2 sin cos 

Aug., 1939] 



W a = W b = 

2 sin 

if a is the angle between the walls. If the stylus has to track a modu- 
lated groove, an additional force St is effective. This force St is equal 
to the difference between the inertia force and the stiffness force. 
For the lower frequencies, it is very nearly equal to the stiffness force 
of the system; for the higher frequencies of the audible spectrum, it 
is a function of the difference between inertia and stiffness. In any 
event, the force St will act horizontally against one wall of the groove, 
as shown in Fig. 2. If the system is stiffness controlled, the stylus will 

-> \H 


FIG. 2. Forces acting on groove. 

bear toward the neutral position ; while if the system is inertia con- 
trolled, the stylus will bear away from the neutral position. 

The new vector diagram for this condition which is shown for the 
stiffness controlled system indicates that the wall force W b becomes 
greater, while the wall force W a becomes smaller. Referring to 
Fig. 2: 

W a 

W b 

sin I o o 


W b 



sin a 

208 A. L. WILLIAMS [J. s. M. P. E. 

Wa + W b ' L 




n a sn a _ . a a 

2 cos ^ cos /3 2 sm ~ cos ~ 

W a +W b = Lcosfi = _^_ = constant 

. a . a 

sin ^ sm ^ 

The arithmetical sum of both wall forces always equals a constant 
(Fig. 2). In fact if stylus pressure and stiffness force are equal, W a 
becomes zero, and W^ will become 1.414 P, if a is 90 degrees, as in 
commercial practice. If the stiffness force should exceed the stylus 
pressure, the stylus will climb up the side of the groove, and the result 
is that the stylus will no longer track the groove. If we assume that 
the system is inertia-controlled, similar conditions exist, except that 
the force St will be oppositely directed. To find the necessary 
stylus pressure P, it is necessary to investigate under which condi- 
tions the wall force St reaches its maximum. Most vibratory pick-up 
systems are stiffness-controlled for the low frequencies until the reso- 
nance point is reached, and are then inertia-controlled for frequen- 
cies above this point. At any frequency the stiffness forces are al- 
ways oppositely directed to the inertia forces. Since this is the case 
it follows that it should be possible to find a balance between the 
higher-frequency forces and the lower-frequency forces. Or, stated 
in another way, it would be desirable that the maximum wall force 
developed by a complex wave-form involving considerable inertia 
forces should be equal to the maximum wall force at a low frequency 
involving the stiffness forces only. In order to develop this thesis, 
it will be necessary to make certain assumptions concerning the char- 
acter of the wave-forms to be engraved. The assumptions will be 
justified later in this paper. 

It is the inherent character of music and speech that we very rarely 
find pure tones. Instead, a complex combination of frequencies is 
ordinarily the case. In recent literature, it has been frequently 
pointed out that the ability of a device to transmit a square wave is, 
in general, an indication of the fidelity of the device. Taking this 
consideration as our fundamental assumption, such a wave-form en- 
graved in a disk will be expressed by 

y = A sin ut -\- ~ A sin 3w/ + -= A sin but + ... - A sin nut (2) 


In analyzing the inertia forces that would be developed upon a stylus 
constrained to track such a groove, we find the second derivative of 
the wave-form with respect to time. Such an operation gives the 
acceleration in the wave-form. 

-j- = u[A cos co/ + A cos Scot -f A cos 5ut + . . . A cos nut] (3) 

2 = u 2 \ A sin ait + 3A sin 3coi + -= A sin 5<at + . . . nA sin nut (4) 
at 2 ]_ J 

To find the point of greatest acceleration for a square wave, it is neces- 
sary to investigate the point at which sin nut = 1, or where nut = w/2. 
For this point, the equation may be written in the following form : 

_ 1" I o A _ """ _1_ C A "7T . . . IT j , > 

5111 O^ + ^^ Sln oZ "T *> A Sln ST "T ' ' UA Sln O (5) 

Zn zn zn -^J 

It can be seen that all the angles in the above equation, with the ex- 
ception of that of the nth member, must give a value smaller than 1. 
We therefore have a more severe case of acceleration if we consider 
such a phase-shift between the components that all of the angles will 
equal ir/2. For this particularly serious case, the acceleration will 
be expressed by equation 6 

jg = - rf(A + 3A + 5A + . . . nA] (6) 

The member on the right is a simple arithmetic series, the sum of 
which is 

And since this sum is an acceleration term, the force of acceleration is 
given by multiplying by the mass factor : 

Where co is, from the above development, the lowest frequency of the 
series ; A , the amplitude at that frequency, and m the mass that is 
being accelerated by the given wave-form. Assuming that the damp- 
ing forces are negligible, the forces supplied by the wall of a groove 
equal the inertia or acceleration force just considered, minus the stiff- 
ness force. The stiffness force is found from the product of the total 
displacement and the coefficient of stiffness. The total displacement 
for this case is given by : 

*-A+A+A+...+l'A (9) 



[J. S. M. p. E. 

Taking 500 cycles as the fundamental, and 10,000 cycles as the upper 
limit of components, the total displacement for the complex wave 
will be 2.133 A and the consequent stiffness force will be given by 
2.133 AK, where K is the coefficient of stiffness. 

Two cases of maximum wall force are conceivable. In one case 
only a low-frequency fundamental note, without overtones, is con- 
sidered. The walls of such a groove have only to overcome the stiff- 
ness of the system, and set up a force depending upon the displace- 
ment of the stylus. The second case is based upon the hypothetical 
wave-form we have analyzed. We have already advanced the theory 
that for best design, the force generated by stiffness at low frequency 
should be equal to the resultant force developed by the complex wave. 


vp Arm 

5 -DiJp*Xemer,t of 3/y/i/S point 
F'^Disp/ of jiqht beam 

alontj Scale H 

B^Angle p/ckup heaa/ rotates trirovyh. 
P-2R9 Angte of incidence eyt/o/s any/e of 

FIG. 3. Method of measuring vertical unevenness of disk motion. 

mA (2*f)*( n * 1 V- 2.133 AK = AK 


Solving this equation for the stiffness factor K, in terms of the mass, 

K = 3.15 X 10 +8 m (11) 

This relation makes it possible to determine the optimum condition 
for inertia and stiffness of the pick-up system. It is obvious that the 
stiffness can be made as small as required but that certain definite 
limitations exist in regard to the inertia of the system. After all, the 
stylus, at least, has to be kept in motion, but even this is not sufficient. 
In the magnetic pick-up, the generator also must suffer motion. 

Aug., 1939] 



Either a piece of magnetic metal must be moved in a magnetic field, 
or a coil of wire must be moved. The crystal pick-up distinguishes 
itself by the fact that the crystal element is a pressure device, and 
generates an electrical voltage proportional to the pressure applied to 
it, and does so with almost negligible motion. A crystal pick-up, 
therefore, can be built with extremely small inertia. The design of a 
pick-up based upon these considerations will be mentioned later. 

The considerations have thus far been limited to the pick-up car- 
tridge itself. Obviously, since this cartridge rides upon a rotating 
disk, at the end of a not inconsiderable arm, it is necessary to examine 
conditions brought in by this pick-up arm and the record itself. It 

7/tc in Seconds -0.773econ(/ > 360' of rctobtt'on of 

FIG. 4. Vertical unevenness of typical disk motion. 

is ideally desirable to provide a very light stylus pressure, but several 
disturbing factors may occur in practical design. While the record 
is moving and the stylus rides in the groove, two points must be con- 
sidered. First, the friction between the stylus point and the record; 
and second, the fact that most of the records are not absolutely true. 
Nor do most commercial turntables run true. In addition to the side 
motion of the pick-up arm, imposed upon it by the helical character 
of the groove, there is also the factor of vertical motion caused by the 
unevenness of the record or turntable. While the stylus is moving 
up, all parts of the cartridge or arm assembly participating in this 
motion experience acceleration and generate force, adding to the 
stylus pressure. In the case of the stylus moving downward, the ac- 



[J. S. M. P. E. 

celeration of the participating parts generates a force directed op- 
positely to the stylus pressure. 

Measurements have been made on commercial disks with the view 
to determining the unevenness and the accelerations caused by such 
unevenness. These measurements were made as follows: A stylus 
riding in a groove, as shown in Fig. 3, had attached to it a small mirror 
that reflected a light-ray upon a screen at a considerable distance from 
the mirror. The motion of the light-point was measured in relation 
to the angular position of the disk. Fig. 4 shows the vertical uneven- 
ness of a typical disk as a function of the angular position, and shows, 
in addition, the resulting vertical velocity and the acceleration of the 

FIG. 5. Showing frictional force on disk. 

stylus caused to track such an unevenness. The point of maximum 
acceleration is indicated, and is 16 cm per sec 2 , assuming a disk ve- 
locity of 78 rpm. These considerations make it desirable to reduce 
the inertia of the pick-up assembly to vertical motion. 

The stylus pressure develops a frictional force between the moving 
disk and the stylus, which acts upon the contact point of the stylus 
in a horizontal direction, and actually has a tendency to lift the 
pick-up arm if the arm is vertically pivoted at a point above the plane 
of the disk. The couple of these friction forces is^, / representing the 
friction forces and / the height of the pivoting point above the plane 
of the disk. To eliminate this couple, the pivoting point must be 

Aug., 1939] 



brought into the plane of the disk (/ = 0). To reduce greatly the 
effect of the couple, however, the pivoting point may be remotely 
located in the horizontal plane from the stylus point, and the distance 
/ may be made as small as possible. Since the lifting forces are ex- 
pressed byfl/r, it follows that the above procedure will greatly reduce 
the moment expressed by the equation. In Fig. 5, the forces and 
their effect are shown. It is obvious that l/r should approach zero. 
These considerations make plain the fact that under certain condi- 
tions, if the disk is uneven, or if the pivoting point is not in the plane 
of the disk, there will be a difference between the static and the dy- 
namic pressure of the stylus. This condition is expressed by the 
following equation : 

T *L! K- 
dt z dt , fl^ (12) 

r* r* r 


FIG. 6. Section of pick-up arm. 

where (I/r^d^s/di' 1 represents the force due to unevenness of the disk, 
and / is the moment of inertia of the whole pick-up assembly and arm. 
(K/r 2 )ds/dt represents the friction forces of the pivoting point, where 
K is the coefficient of bearing friction, assuming that these forces are 
proportional to the velocity, fl/r represents the lifting forces caused 
by friction. 

In Fig. 6 is shown a section view of a pick-up arm designed after 
the above considerations. It consists of two sections, one of which 
has a horizontal pivoting point where the pick-up arm is mounted, 
and forms a bearing for the other part, which holds the cartridge. 
This first part is made heavy, so as to give the pick-up arm a consider- 
able moment of inertia in the horizontal plane. This helps to bring 
the arm resonance down to a very low frequency, where it is not a 
problem. The other section, holding the cartridge, and pivoting 



[J. S. M. P. E. 

vertically in the first section, is very light, to reduce the moment of 
inertia in the vertical plane. This makes the pick-up relatively in- 
sensitive to the unevenness of the record and turntable surfaces. 
Fig. 7 shows a finished arm, designed after these considerations. 

It has proved quite a problem to determine the maximum stylus 
pressure that can be used without causing apparent wearing of soft 
disk materials. In connection with the investigation of the relation 
between stylus pressure and friction, we come to some very interest- 
ing conclusions. In Fig. 8 the friction forces are shown as functions 
of the dynamic pressure. Curve A shows this relation for an un- 
modulated groove. Curve B is taken from a 200-cycle groove modu- 
lated at an amplitude of 0.0004 inch and curve C is from 2500 cycles 
at the same amplitude. Curve A shows a proportionality between 

FIG. 7. Photograph of pick-up arm. 

pressure and friction up to around 60 grams, and the curve then bends 
and starts to rise more rapidly beyond this point. Curves B and C 
depart from linearity at about 25 grams' pressure. Microscopic ex- 
amination has shown that no excessive wear takes place on the record 
grooves at stylus pressures below the inflection point in the curve. 
Beyond that point, however, noticeable wear takes place. From 
these observations, we conclude that a groove cut in nitrate material, 
and modulated at peak amplitude with a fairly high frequency will 
not stand more than 25 grams' pressure. 

When it is considered that most of the conventional pick-ups are 
still engineered to operate at stylus pressures in excess of l l / z to 2 
ounces, it is easy to see that permanence in nitrate records is still a 
pleasant fantasy. Using the improved Brush pick-up, however, ace- 
tate recordings have been reproduced many times, without the slight- 

Aug., 1939] 



est depreciation in quality. The stylus pressure in these tests has 
always been less than 25 grams. 

The method which has been used for measuring the friction force 
developed between the stylus tip and the groove of a record is as 
follows : The requirement is to measure the drag of the disk upon the 
pick-up assembly, in a line tangent to the groove being tracked, and 
in the plane of the disk. Since it is required that the dynamic stylus 
pressure being used should be known as accurately as possible, the de- 
vice must combine low bearing friction and high sensitivity, and the 
vertical pivot must be in the plane of the disk being tested. 


Sty/in Pr 


FIG. 8. Frictional forces in terms of dynamic pressure. 

The apparatus is shown in Fig. 9. The pick-up head A is mounted 
in arm B. Arm B, constructed of a light alloy, is pivoted at K, on a 
bearing allowing free vertical motion of the arm. Arm F is secured 
solidly to arm C, and springs G are fixed to the pick-up arm B and 
connected together at their other end by a cord running over pins L 
and passing through a clamp at E. The pressure of the stylus on the 
disk may be adjusted by changing the relative tensions in the springs 
G, and such adjustments are held by clamping the cord connecting 
the two springs, at R. D is an arm pivoted at H in a ball bearing 
allowing horizontal motion and having affixed to its free end the whole 
assembly heretofore mentioned, at bearing M. Bearing M allows 
the sidewise motion required for the apparatus to track. It will be 



[J. S. M. P. E. 

noted that the assembly hangs from M as a pendulum bob, but for the 
relatively small time required to take a reading with the apparatus, 
the pendulum will be only very slightly displaced from its neutral po- 
sition, and the error introduced by this factor is of vanishing impor- 

Connected to arm D by means of a light thread is the spring bal- 
ance P, constructed after the idea of a Jolly balance, so that the read- 
ing is taken after the arm D is returned to a neutral position by the 
force of the balance. This procedure tends to cancel out errors that 
would be introduced if the arm were at different points for different 

FIG. 9. Apparatus for measuring friction between stylus tip and groove. 

friction forces. The set-up measures friction force directly in grams. 

The stylus pressure was measured with a spring balance. The 
balance was constructed with an eye to the special requirements of 
the measurements to be made, but is essentially a sensitive spring 
balance, calibrated in grams. A saddle from the spring holds the 
stylus, and the tension on the spring is increased until the saddle just 
lifts the stylus free from the disk surface. 

Some of the design features used in the Brush pick-up will now be 
taken up. In Fig. 10 the disk reproducer cartridge is shown. The 
sapphire stylus is held in a hollow tubing which is connected to a drive 

Aug., 1939] 



wire. This wire is fastened to the crystal element. The wire itself 
is held in bearings which permit torsional motion. The stylus point 
is forced by the grooves to describe an extremely small section of the 
periphery of a circle, so small that the arc is essentially equal to the 
chord. While doing so, the wire is twisted, which develops a torque 
pressure on the crystal element. The crystal element generates a 
voltage proportional to this pressure, and it may be again pointed out 
here that this voltage will at any time be proportional to the displace- 
ment of the stylus tip and not proportional to the velocity. 

We do not have to concern ourselves with details about the crystal 
element because this element neither adds substantially to the stiff- 
ness nor to the inertia of the vibratory system consisting of stylus and 
tubing and wire drive. Due to the fact that the wire moves only in 
torque, it adds negligibly to the 
inertia of the system, and repre- 
sents the force of stiffness in our 
record reproducer. The inertia 
of the system is concentrated 
mostly in the sapphire point and 
the extremely thin-walled tubing 
that holds the sapphire. It has 
already been pointed out that the 
mass of the system is the de- 
terming factor for the stiffness. 
The mass must be made as small 
as possible, and in this connection 
a signal step has been taken for- 
ward in the design explained above. The entire structure of this 
assembly is contrasted with that of a conventional chromium stylus 
in Fig. 11. 

The resonance frequency of the crystal assembly has been raised 
still higher above the audible range, oil damping has been added to 
the crystal, the inertia of the stylus has been greatly reduced, and the 
stiffness of the drive wire has been readjusted to cooperate properly 
with this reduced inertia. The effective inertia of the stylus as- 
sembly may be reduced to 2.22 X 10~ 6 gm sec 2 cm" 1 at the stylus 
point. The stiffness, therefore, after formula 10, should be equal to 

K = 3.15 X 10 8 w = 700 gm/cm 

By actual measurement, the assembly has a stiffness of approximately 
800 gm/cm. 

FIG. 10. Disk reproducer cartridge. 



[J. S. M. P. E. 

For purposes of testing a pick-up, ;it is extremely desirable that a 
single frequency should be found that would simulate the conditions 
imposed by the complex- wave hypothesis. Since 10,000 cycles per 
second is assumed to be the highest frequency to be reproduced, this 
frequency is selected as the single frequency to simulate the complex 
wave-form. The total wall force generated by the complex wave- 
form is given by : 

F = 

- 2.133 AK 


Setting this equal to the wall force of a 10,000-cycle note and solving 
for the amplitude of this 10,000 cycle note : 

n* - K] (14) 


It follows that the hypothetical wave- 
form will generate a wall force equal 
to that generated by a 10,000-cycle 
pure tone, recorded at an amplitude 
0.0973 A, where A is the amplitude 
of the fundamental frequency of the 
hypothetical wave. 

Even assuming, however, that the 
wall forces are negligible, that in itself 
will not assure tracking of the 10,000- 
cycle note. It is required that the 
stylus tip shall follow the undulations 
of the wave, but obviously this is 

possible only if the radius of curvature of the stylus tip is less than 
that of the wave-form being tracked. It is necessary, therefore, to 
determine the radius of greatest curvature of the hypothetical wave- 
form. The radius of curvature is expressed by the following equation : 


FIG. 11. Sapphire stylus as- 
sembly, contrasted with conven- 
tional chromium stylus. 


It has been shown previously in this paper that for the hypothetical 
wave form : 


37 = o> (A cos wt + A cos 3o)t + . . . + A cos n0 



? = W 2 (4 sin ^1 + 3A sin 3wt + . . . + nA sin wo>/) 
a/ 2 

The greatest curvature will be at the point where the greatest accel- 
eration takes place, or where sin ut = sin 3 co/ = . . . = sin nut = 1 . 
In this case dy/dt = and we find that 

p = 


It must be possible to find a single frequency of the nth component 
that will have the same curvature. This frequency is expressed by : 

By setting both curvatures equal : 

From this we find that A* = 0.276 A, assuming 10,000 cycles is the 
highest frequency in the complex wave- form, or to express the same 
idea in different terms, if A is the amplitude of the fundamental fre- 
quency of the hypothetical wave, the amplitude of a sine wave of nth 
order which has the same curvature as the complex wave-form must 
be approximately Y 4 A . 

It will be of interest to mention a specific example where the for- 
mula for the radius of curvature is used to determine what minimum 
disk diameter is necessary to reproduce the hypothetical wave. The 
formula for the radius of curvature can be reduced to the following 
simple equation : 

1.69r 2 



where r is the radius of the disk at the groove being considered; / is 
the frequency of the signal, assumed to be a sine wave ; and A is the 
amplitude of the signal at that frequency. The constant 1.69 is 
based upon a disk speed of 78 rpm (0.308 is the constant for 33 Vs 

The conventional tracking stylus has a point radius of 0.0025 inch. 
Fig. 1 shows such a stylus resting in a 90-degree stylus cut, and it will 
be seen that the curvature in contact with the groove walls will have 
a radius of 1 / 2 the length of the chord connecting the contact points 
of the groove and the stylus. This value will always be less than the 



[J. S. M. P. E. 

radius of the stylus point, and for a 90-degree cut, is given as 0.707 
the stylus point radius. Returning, now, to our example: The fre- 
quency of the highest component in the hypothetical wave is 10,000 
cycles. The conventional tracking stylus has a point radius of 
0.0025 inch and from above considerations, the radius of the contact 
circle is 0.707 X 0.0025 or 0.00177 inch. Therefore, in formula 19, 
p must be 0.00177 inch. The amplitude of the equivalent 10,000- 
cycle wave, from previous considerations is 0.276 A, and assuming 
that 0.0002 inch is the fundamental amplitude A, the amplitude to be 
used in the formula is 0.0002 X 0.276 = 0.0000552 inch. Solving the 
formula for r shows that at 78 rpm the minimum disk radius that will 
reproduce the hypothetical wave-form is 2.4 inches. 










"onstant drr 



Y' CL 




. \ 



c/ y 






^ , 


kx ^ 

x x 






j : 





^ '" 






x > 








? . 








FIG. 12. (Solid curve) Energy spectrum (H. Fletcher). 

Since we have discussed, to a great extent, the effects of a complex 
wave-form upon a pick-up, and have used a specific example of such 
a wave-form from which to draw our conclusion, it is, of course, very 
desirable to justify the assumptions inherent in these considerations. 
Such justification, fortunately, can be found from an analysis of the 
energy spectrum for speech and music. 

Very pertinent information of this nature is supplied by Fletcher, * 
as shown in Fig. 12. This curve has a definite peak at about 350 
cycles, and falls off both above and below this point. As indicated in 
this figure, an envelope curve is shown that is flat from to 500 cycles, 
and falls thereafter in such a way that the product of the amplitude 

Aug., 1939] 



and frequency is constant. While it can be seen that this envelope 
curve somewhat exaggerates the condition, it certainly shows a con- 
dition that is rather more severe than measurements indicate. Tak- 
ing this envelope curve as the basis for consideration, and assuming 
that the turnover point of this curve represents the fundamental 
frequency, we find that our hypothetical wave built up by this fun- 
damental frequency and all odd harmonics having amplitudes in- 
versely proportional to their order (Eq. 2) is expressed by the dotted 
envelope curve. It is almost inconceivable that the wave-form would 
contain all the harmonics, in the relationship expressed by the en- 
velope curve. It is somewhat more in accordance with the law of 
greatest probability to assume that only every other harmonic is 

Ok to 


FIG. 13. Frequency response and harmonic content of repro- 
ducer cartridge. 

present, and this is what has been done in the hypothetical wave. 
Making this assumption we more closely approach reasonable ex- 
pectancy for speech and music. It will now be seen that these con- 
siderations are based upon an envelope spectrum which would be 
found recorded upon a disk if that record were cut with a constant- 
amplitude device. 

Very successful records have been cut with an amplitude of 0.0002 
inch maximum displacement. This amplitude is approximately 20 
db down with respect to the average amplitude of a Victor record 
for the low frequencies, but it is approximately 10 db higher for 
10,000 cycles. Since the scratch noise of a record is in the higher 
range, constant-amplitude recording means that the scratch noise is 
considerably reduced. It has the other advantage that the grooves 

222 A. L. WILLIAMS [j. s. M. P. E. 


can be cut closer to each other, leading to a longer possible playing 
time of a record of the same diameter. 

Using a crystal cutter and a crystal pick-up, it is even more natural 
to consider constant-amplitude recording. With the crystal as a 
driver in a record cutter it is easy to produce a motion that is propor- 
tional to the voltage applied to it. Conversely, any crystal element 
generates a voltage corresponding to pressure applied to it. The 
Brush pick-up can be used without any equalization whatever for the 
reproduction of a record cut with constant-amplitude characteristics, 
so long as it is terminated into a high-impedance input. All that has 
to be done, therefore, is to connect the pick-up to the grid of the first 
stage of the amplifier. Provisions are made for either push-pull or 
single-channel input. 

In Fig. 13 the frequency response and harmonic content of this 
cartridge are shown. In the frequency range of 30 to 10,000 cycles, the 
amplitude variation is less than 2 db, and the total harmonic con- 
tent is less than 1.5 per cent. It should be pointed out that the dis- 
tortion curves are for the complete system, including the oscillator, 
amplifier, cutter, disk material, and reproducer. It is reasonable to 
believe, therefore, that the actual amount of distortion due to the 
pick-up is considerably less than that shown in these curves. 

Measurement places the resonance frequency of the crystal as- 
sembly in the neighborhood of 24,000 cycles, which is well above the 
audible range. Even so, sufficient damping is used so that the reso- 
nance peak at 24,000 cycles is less than 10 db. 

By the use of the "off-set" principle, of which much has been said 
in the literature of the past few years, maximum tracking error on a 
12-inch disk with a 12-inch arm is held to approximately 1 degree, a 
negligible error. 


1 FLETCHER, H.: "Some Physical Characteristics of Speech and Music," Bell 
Syst. Tech. /., July, 1931. 


MR. KELLOGG: How is the shaft pivoted at the upper end of the stylus bar? 

MR. WILLIAMS: The beryllium bronze rod is supported in metal bearings. 
Rubber could be used. 

MR. KELLOGG : Your diagram indicates a spherical reproducing stylus resting 
entirely on the side of the groove. Is it my understanding that for lateral disk 
recording steel cutters are sharpened and that the sapphires are rounded? Of 


course, in all standard recordings the round-ended cutting stylus is used with sub- 
stantially the same radius as the reproducer. 

MR. WILLIAMS: It is our practice to use 0.0023 inch for the cut and 0.0025 
inch for the pick-up. The diagram showed 90 degrees for simplicity, which is a 
little broader than the angle used. 

MR. KELLOGG: Have you found using the same size superior, or have you 
found any difference? 

MR. WILLIAMS: I believe it is definitely superior. Of course, with the steel 
needle it is not possible, as the needle will wear. We do not expect any wear on 
the stylus and very little on the record. We always expect a tight fit. You can 
not have good reproduction unless you do have a tight fit. 

MR. CRABTREE: What improvement in quality are we to expect by the use 
of this pick-up? In other words, what is there on the record that we can not get 
with existing pick-ups? 

MR. WILLIAMS: Most of the records do not have the higher frequencies. 
One of the reasons for this, I believe, is that when they do record at high fre- 
quencies, the large mass of the moving parts of the ordinary pick-up due to the 
size of the stylus, etc., wears off at these higher frequencies very quickly, causing 
an increase in surface noise. If this record can be brought out recorded with 
constant amplitude, it will be possible to get a quarter on an hour's playing time 
with a 12-inch record using the same amplitude as we were here and going up to 
10,000 cycles without any trouble. 



The editors present for convenient reference a list of articles dealing with subj< 
cognate to motion picture engineering published in a number of selected jour 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C. 

American Cinematographer 

20 (June, 1939), No. 6 

Problems Faced by Labs in Australia (pp. 246-247) E. HUSE 
Television Highlights Engineers' Convention (pp. 254- 

257) W. STULL 

Improved Wild Cinemotor Developed (pp. 259-260) R. N. HAYTHORNE 
Filtering Arcs for Matching Quality in Monochrome 

(pp. 269-270) C. B. LANG, JR. 

British Journal of Photography 

86 (May 5, 1939), No. 4122 
Progress in Colour (pp. 279-281) 

86 (May 12, 1939), No. 4123 
Progress in Colour (pp. 294-295) 

Educational Screen 

18 (May, 1939), No. 5 
Motion Pictures Not for Theaters (pp. 153-156) 

Pt. 9 A. E. KROWS 

International Photographer 

11 (May, 1939), No. 4 

The BNC Mitchell Silent Camera (pp. 7-9) S. POLITO 

Fundamental Photographic Physics (pp. 12-14) Pt. 3 D. HOOPER 
Grip Equipment (pp. 14-15) G. M. HAINES 

International Projectionist 

14 (May, 1939), No. 5 
Many Important Changes in N. F. P. A. Projection 

Room Regulations (pp. 15, 27-28) 
Process Projection Specifications. Pt. 1. Report of 
Research Council, Academy of Motion Picture Arts 
and Sciences (pp. 17-18, 24-27) 

Film Preservative Tests. Abstract of a report on Re- 
lease Print Quality Committee Research Council, 




Academy of Motion Picture Arts & Sciences 
(pp. 20-21) 

New Erpi Mirrophonic 'Master' Theater Sound System 
(pp. 21-22) 


21 (Apr., 1939), No. 4 

Messungen an Bildschirmen. (Measurements on Pro- 
jection Screens) (pp. 89-92) 
Der vollkommene plastische Films (The Perfect 

Stereoscopic Film) (pp. 92-97) Pt. 2 
Projektion mit Quecksilberlicht (Projection with a 

Mercury Light and Discussion) (pp. 98-101) 
Italienische und englische Filmatelier-Betriebe (Italian 
and English Film Studios) (pp. 101-106) 

21 (May, 1939), No. 5 

rbeitung des Sicherheitsfilmes, 35 Mm. (Work on 
35-Mm. Safety Film) (pp. 116-118) 
Din neues Material fur Tonaufzeichnung in Zachen- 
schrift (A New Material for Variable Width Sound 
Recording) (pp. 118-122) 

irtgleichungen (Comparison Tables) (pp. 122-126) 
Pt. I 

Fnvergangliches Lichtbild und Metallfilm (Imperish- 
able Pictures and Metal Films) (pp. 127-129) 
r ersuch der DKG Betr. die Negativ-Entwicklung in 
Kopieranstalten (Investigation of the DKG Nega- 
tive Development in Printing Establishments) (pp. 

lotion Picture Herald (Better Theaters Section) 

135 (May 27, 1939), No. 8 

The Advantages of the Smaller Image for Color Films 
(pp. 38-39) 

Photographic Society of America, Journal 
5 (May, 1939), No. 2 

16-Mm. Sound Films by Direct Recording (pp. 16-19) 
Recent Developments in Photographic Objectives (pp. 










Officers and Committees in Charge 

E. A. WILLIFORD, President 

S. K. WOLF, Past-President 

W. C. KUNZMANN, Convention Vice-P resident 

J. I. CRABTREE, Editorial Vice-P resident 

D. E. HYNDMAN, Chairman, Atlantic Coast Section 
J. HABER, Chairman, Publicity Committee 

S. HARRIS, Chairman, Papers Committee 

H. GRIFFIN, Chairman, Convention Projection 

E. R. GEIB, Chairman, Membership Committee 

Reception and Local Arrangements 

D. E. HYNDMAN, Chairman 







Registration and Information 

W. C. KUNZMANN, Chairman 



C. Ross 





Hotel and Transportation 

J. FRANK, JR., Chairman 


J. HABER, Chairman 

P. A. McGuiRE 

P. A. McGuiRE 


Convention Projection 

H. GRIFFIN, Chairman 






Officers and members of Projectionists Local 306, IATSE 

Banquet and Dance 

A. N. GOLDSMITH, Chairman 





Ladies 1 Reception Committee 

MRS. O. F. NEU, Hostess 






Headquarters. The headquarters of the Convention will be the Hotel Pennsyl- 
vania, where excellent accommodations have been assured, and a reception suite 
will be provided for the Ladies' Committee. 

Reservations. Early in September room reservation cards will be mailed to 
members of the Society. These cards should be returned as promptly as possible 
in order to be assured of satisfactory accommodations. The great influx of visi- 
tors to New York, because of the New York World's Fair, makes it necessary to 
act promptly. 

Hotel rates. Special per diem rates have been guaranteed by the Hotel Penn- 
sylvania to SMPE delegates and their guests. These rates, European plan, will 
be as follows: 

Room for one person $ 3 . 50 to $ 8 . 00 

Room for two persons, double bed $ 5.00 to $ 8.00 

Room for two persons, twin beds $ 6.00 to $10.00 

Parlor suites: living room, bedroom, $12.00, $14.00, and 
and bath for one or two persons $15.00 

Parking. Parking accommodations will be available to those who motor to 
the Convention at the Hotel Fire Proof Garage, at the rate of $1.25 for 24 
hours, and $1.00 for 12 hours, including pick-up and delivery at the door of the 

228 1939 FALL CONVENTION [j. s. M. P. E. 

Registration. The registration desk will? be located on the 18th floor of the 
Hotel at the entrance of the Banquet Room on the ballroom floor where the 
technical sessions will be held. Express elevators from the roof will be reserved 
for the Convention. All members and guests attending the Convention are 
expected to register and receive their badges and identification cards required for 
admission to all the sessions of the Convention, as well as to several de luxe motion 
picture theaters in the vicinity of the Hotel. 

Technical Sessions 

The technical sessions of the Convention will be held in the Banquet Room on 
the ballroom floor of the Hotel Pennsylvania. The Papers Committee plans to 
have a very attractive program of papers and presentations, the details of which 
will be published in a later issue of the JOURNAL. 

Luncheon and Banquet 

The usual informal get-together luncheon will be held in the Roof Garden of the 
Hotel on Monday, October 16th. 

On Wednesday evening, October 18th, will be held the Semi-Annual Banquet 
and Dance, also in the Roof Garden of the Hotel. At the banquet the annual 
presentation of the SMPE Progress Medal and the Journal Award will be made, 
and the officers-elect for 1940 will be introduced. 


Motion Pictures. At the time of registering, passes will be issued to the dele- 
gates of the Convention admitting them to several de luxe motion picture theaters 
in the vicinity of the Hotel. The names of the theaters will be announced later. 

Golf. Golfing privileges at country clubs in the New York area may be ar- 
ranged at the Convention headquarters. In the Lobby of the Hotel Pennsylvania 
will be a General Information Desk where information may be obtained regard- 
ing transportation to various points of interest. 

Miscellaneous. Many entertainment attractions are available in New York to 
the out-of-town visitor, information concerning which may be obtained at the 
General Information Desk in the Lobby of the Hotel. Other details of the enter- 
tainment program of the Convention will be announced in a later issue of the 

Ladies' Program 

A specially attractive program for the ladies attending the Convention is being 
arranged by Mrs. O. F. Neu, Hostess, and the Ladies' Committee. A suite will 
be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. Further details will be published in a succeeding 
issue of the JOURNAL. 

New York World's Fair 

Members are urged to take advantage of the opportunity of combining the 
Society's Convention and the New York World's Fair on a single trip. Informa- 
tion on special round-trip railroad rates may be obtained at local railroad ticket 

Aug., 1939] 1939 FALL CONVENTION 229 

offices. Trains directly to the Fair may be taken from the Pennsylvania Station, 
opposite the Hotel: time, 10 minutes; fare, 10^. Among the exhibits at the 
Fair are a great many technical features of interest to motion picture engineers. 

Points of Interest 

Headquarters and branch offices of practically all the important firms engaged 
in producing, processing, and exhibiting motion pictures and in manufacturing 
equipment therefor, are located in metropolitan New York. Although no special 
trips or tours have been arranged to any of these plants, the Convention provides 
opportunity for delegates to visit those establishments to which they have entree. 
Among the points of interest to the general sightseer in New York may be listed 
the following : 

Metropolitan Museum of Art. Fifth Ave. at 82nd St.; open 10 A.M. to 5 P.M. 
One of the finest museums in the world, embracing practically all the arts. 

New York Museum of Science and Industry. RCA Building, Rockefeller Cen- 
ter; 10 A.M. to 5 P.M. Exhibits illustrate the development of basic industries, 
arranged in divisions under the headings food, industries, clothing, transportation, 
communications, etc. 

Hayden Planetarium. Central Park West at 77th St. Performances at 1 1 A.M., 
2 P.M., 3 P.M., 4 P.M., 8 P.M., and 9 P.M. Each presentation lasts about 45 minutes 
and is accompanied by a lecture on astronomy. 

Rockefeller Center. 49th to 51st Sts., between 5th and 6th Aves. A group of 
buildings including Radio City Music Hall, the Center Theater, the RCA Build- 
ing, and the headquarters of the National Broadcasting Company, in addition to 
other interesting general and architectural features. 

Empire State Building. The tallest building in the world, 102 stories or 1250 
feet high. Fifth Ave. at 34th St. A visit to the tower at the top of the building 
affords a magnificent view of the entire metropolitan area. 

Greenwich Village. New York's Bohemia; a study in contrasts. Here are 
located artists and artisans, some of the finest homes and apartments, and some 
of the poorest tenements. 

Foreign Districts. Certain sections of the city are inhabited by large groups of 
foreign-born peoples. There is the Spanish section, north of Central Park; the 
Italian district near Greenwich Village; Harlem, practically a city in itself, num- 
bering 300,000 negroes; Chinatown, in downtown Manhattan; the Ghetto, the 
Jewish district; and several other such sections. 

Miscellaneous. Many other points of interest might be cited, but space permits 
only mentioning their names. Directions for visiting these places may be ob- 
tained at the Convention registration desk: Pennsylvania Station, Madison 
Square, Union Square, City Hall, Aquarium and Bowling Green, Battery Park, 
Washington Square, Riverside Drive, Park Avenue, Fifth Avenue shopping dis- 
trict, Grand Central Station, Bronx Zoo, St. Patrick's Cathedral, St. Paul's 
Chapel, Cathedral of St. John the Divine, Trinity Church, Little Church Around 
the Corner, Wall St. and the financial district, Museum of Natural History, 
Columbia University, New York University, George Washington Bridge, Brook- 
lyn Bridge, Triborough Bridge, Statue of Liberty, American Museum of Natural 
History, Central Park, Metropolitan Museum of Art, and Holland Tunnel. 



At a meeting held on June 21st at the RCA Photophone Studios, New York, 
an elaborate demonstration of equipment, with descriptive paper, was given by 
Messrs. A. Goodman, R. J. Kowalski, W. F. Hardman, and E. S. Stanko of the 
RCA Manufacturing Company, Camden, N. J. The subject of the paper was 
"Safeguarding Theater Sound Reproduction with Modern Test Instruments." 
The emphasis of the presentation was on the instrumental means used by field 
engineers in servicing sound reproduction equipment. Among the instruments 
described were a cathode-ray oscillograph for plotting upon the fluorescent screen 
the response characteristic of an amplifier system; an instrument for plotting 
rapidly, with the aid of a warble-frequency film, the acoustic response of an audi- 
torium; an extremely sensitive bridge for measuring electrical constants of re- 
producing circuits; a sound-level meter; and other instruments of great aid to 
the installation and service engineer. 


At a meeting held at the General Service Studio at Hollywood, on June 27th, a 
special introduction and demonstration of the Vocoder was given through the 
courtesy of Electrical Research Products, Inc., and Bell Telephone Laboratories. 
The technical discussion and demonstration were made by Mr. H. Dudley of Bell 
Telephone Laboratories. The Vocoder, or voice analyzer-synthesizer, is some- 
what similar to the Voder, which is on demonstration at the World's Fair. Its 
main difference is that the control is by a speaker's voice rather than by manually 
operated keys. Sounds are first separated into frequency bands, any of which 
may be treated in various ways at the will of the operator before recombining to 
produce some desired effect. 

With this instrument, means are provided for changing speech or music in in- 
numerable ways, by altering the tonal quality or characteristics, by varying or 
reversing the inflection, by raising or lowering the pitch, by adding other effects 
such as tremolo, to simulate the quavering voice of an aged person; and by modu- 
lating music or sound effects. The change in voice, as well as the creation of 
rather weird and inhuman, but still intelligible speech, has suggested the possi- 
bility of using the Vocoder principle in motion picture work, particularly for 
cartoons and in regular dubbing and sound-effects recording. 


A meeting of the Sub-Committee on Fire Hazards was held at the office of the 
Society on June 28th, the principal subject of discussion being the heating of 
projection rooms. 


A meeting of the general committee was held on July 20th, at the Paramount 
Building, New York, N. Y., at which time reports of the various sub-committees 
were considered for inclusion in the report of the Committee for the approaching 
Fall Convention. The Committee is pleased to announce that the proposed re- 
visions of the NFPA Regulations for Handling Nitrocellulose Motion Picture 
Film were adopted, in their principal features, at the recent NFPA meeting at 
Chicago. A new edition of the Regulations is being prepared by the NFPA. 


At a recent meeting of the Admissions Committee at the General Office of the 
Society, the following applicants for membership were admitted to the Associate 


290 Horns Rd., Perry, Ga. 

Ilford, Essex, England. HOLMES, O. J. 

BLOCK, O. 141 6 W. Clay St., 

610 Trinity Ave., Richmond, Va. 

Bronx, N. Y. NICKEL, F., JR. 

CURRIE J E 229-4th Ave., 

356 W. 44th St. New York ' N " Y ' 

New York, N. Y. PARISEAU, S. M. 

1584 W. Washington Blvd., 

DOLLMAN, S. C. Los Angeles , Calif. 

River Rd., STACE> p N 

Bound Brook, N. J. Wellington, New Zealand. 


2202 80th St., 210 Butler Ave., 

Brooklyn, N. Y. Buffalo, N. Y. 


These films have been prepared under the supervision of the Projection 
Practice Committee of the Society of Motion Picture Engineers, and are 
designed to he used in theaters, review rooms, exchanges, laboratories, 
factories, and the like for testing the performance of projectors. 

Only complete reels, as described below, are available (no short sections 
or single frequencies). The prices given include shipping charges to all 
points within the United States; shipping charges to other countries are 

35-Mm. Visual Film 

Approximately 500 feet long, consisting of special targets with the aid 
of which travel-ghost, marginal and radial lens aberrations, definition, 
picture jump, and film weave may be detected and corrected. 

Price $37.50 each. 

16-Mm. Sound-Film 

Approximately 400 feet long, consisting of recordings of several speak- 
ing voices, piano, and orchestra; buzz-track; fixed frequencies for focus- 
ing sound optical system; fixed frequencies at constant level, for de- 
termining reproducer characteristics, frequency range, flutter, sound- 
track adjustment, 60- or 96-cycle modulation, etc. 

The recorded frequency range of the voice and music extends to 6000 
cps.; the constant-amplitude frequencies are in 11 steps from 50 cps. to 
6000 cps. 

Price $25.00 each. 

16-Mm. Visual Film 

An optical reduction of the 35-mm. visual test-film, identical as to 
contents and approximately 225 feet long. 
Price $25.00 each. 







Volume XXXin September, 1939 



Flicker in Motion Pictures L. D. GRIGNON 235 

The Work of the Process Projection Equipment Committee of 

the Research Council, Academy of Motion Picture Arts and 

Sciences A. F. EDOUART 248 

A Cardioid Directional Microphone 


Characteristics of Modern Microphones for Sound Recording. 

F. L. HOPPER 278 

The Class A-B Push-Pull Recording System 


RCA Aluminate Developers J. R. ALBURGER 296 

The Present Technical Status of 16-Mm. Sound-Film 

J. A. MAURER 315 

The Fluorescent Lamp and Its Application to Motion Picture 

Studio Lighting G. E. INMAN AND W. H. ROBINSON 326 

Presidential Address ; 1939 Spring Convention . E. A. WILLTFORD 336 

Current Literature 341 

1939 Fall Convention at New York, N. Y., October 16th-19th, 

Inclusive 343 

Society Announcements 347 





Board of Editors 
J. I. CRABTREE, Chairman 




Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1939, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


** President: E. A. WILLIFORD, 30 East 42nd St., New York, N. Y. 
** Past-President: S. K. WOLF, RKO Building, New York, N. Y. 
** Executive Vice-President: N. LEVINSON, Burbank, Calif. 

* Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
** Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 

* Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 
** Convention Vice-P resident: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St., New York, N. Y. 

* Treasurer: L. W. DAVEE, 153 Westervelt Ave., Tenafly, N. J. 

** M. C. BATSEL, Front and Market Sts., Camden, N. J. 

* R. E. FARNHAM, Nela Park, Cleveland, Ohio. 

* H. GRIFFIN, 90 Gold St., New York, N. Y. 

* D. E. HYNDMAN, 350 Madison Ave., New York, N. Y. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* S. A. LUKES, 6427 Sheridan Rd., Chicago, 111. 

** H. G. TASKER, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1939. 
** Term expires December 31, 1940. 


Summary. Flicker in motion pictures has been receiving attention ever since the 
beginning of the art, and most of the sources of this defect have been minimized, if not 
eliminated, by technical accomplishments. The paper constitutes a qualitative review 
of the now prevalent sources of flicker, presenting some new concepts, emphasizing the 
sources of major importance at the present time, and reporting on two investigations 
made on the problem. Flicker and "registration jump" are differentiated, and the 
latter, which is really a separate problem, is not considered. Some data are presented 
to indicate the magnitude and characteristics of the flicker effect. 

Constant efforts have been directed in the technical branches of 
motion picture production and exhibition toward the removal of 
effects which make the mechanical processes in pictures obvious to 
the observers and detract thereby from the realism and entertain- 
ment value. Aside from features such as camera angle, lighting, 
sets, backgrounds, sound, etc., two completely mechanical effects in 
pictures can cause serious loss of entertainment value. These two 
are flicker and registration. 

This paper does not propose to discuss registration; therefore it is 
necessary to differentiate this effect from that of flicker. Briefly, 
registration is an irregularity in the position of successive picture 
frames on the film or screen. Flicker is an irregularity between 
successive frames in the total amount of reflected light from the screen, 
other than that purposely created, from a given scene. 

Flicker still is an important problem in the industry although the 
serious defects are intermittent in nature. Flicker is due not only 
to the frame frequency (24 per second) but also is the result of other 
variations superimposed upon the frame frequency. This latter 
effect can be considered in the same light as flutter in sound recording 
and reproducing. This paper will lay the greatest stress on these 
harmful superimposed variations. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
20, 1939. 

'* Paramount Pictures, Inc., Hollywood, Calif. 


236 L. D. GRIGNON [j. s. M. p. E. 

Consider now the many sources of flicker, and group them as follows 
for later consideration : 

(.4) Original Photography 

(1) Set lighting 

(2) Negative film 

C?) Irregular camera motion including motor system 
(4) Development 
() Printing 

(1) Lamp irregularity 

(2) Positive 

(3) Printer motion 

(4) Development 
(Q Projection 

(1) Arc flicker 

(2) Intermittent shutter 

(5) Projector mechanics 
(D) Background Projection 

(1) All of A 

(2) All of B 

(3) All of C 

With so many possible sources of flicker it is very easy to under- 
stand how flicker may easily occur. Also, although each of the above 
might be small in absolute value, in instances when two or more 
occur at the proper frequencies and phase relationships the effect 
becomes pronounced. This likewise accounts for the difficulty in 
tracing, separating, and minimizing the major sources. 

Two analyses and investigations made at the Paramount Studios in 
Hollywood disclosed four important facts : 

First, considerable change in reflected light can be tolerated by the observer 
provided this change occurs at random intervals which are not closely spaced or 
of excessive duration. The moment the light change becomes cyclic the amount of 
tolerable difference decreases sharply to a surprisingly small value. 

Second, the rate at which the cyclic flicker occurs determines the amount of dis- 
turbance to the observer. No accurate determinations of this fact have been 
made. However, the rate of maximum disturbance appears to be between 6 and 8 
cycles per second. Fig. 1 shows an approximate curve representing the apparent 
disturbing effect versus the rate of flicker. 

Third, the change in transmission for perceptible periodic flicker occurring at 
the greatest disturbing rate of 6 to 8 cycles is about 3 per cent. The greater the 
change in transmission the greater the effect. 

Fourth, the disturbing effect is related to the amount of light. The greater the 
intensity the more obvious the defect. 

Sept., 1939] 



We shall now discuss the various sources of flicker, some briefly and 
others in more detail. 

(A) Original Photography. The first cause of flicker in this 
group occurs in the set lighting. The intensity changes of incandes- 
cent lamps are of a relatively slow and random nature, and cause 
changes in the average brilliance of the scene and are dependent 
upon the regulation and stability of the power supply. Arc lamp 
flicker is more likely to be cyclic and therefore of a more serious 
nature. Arc lamp flicker generally resolves into slow periodic 
changes such as line-voltage and carbon-rotation effects and very 
fast random fluctuations. The very rapid fluctuations cause the 
most trouble in background process work while the slow variations 

1 .O 


^ x 












FIG. 1. Apparent disturbing effect of cyclic flicker iw. the rate of flicker. 

can cause serious trouble on split-screen shots. This is acknowledged 
and must be solved by the lamp and carbon manufacturers. 

Negative film is known to have random changes in sensitivity, and 
some stocks have cyclic changes occurring at a rate of one cycle in 
7 to 12 seconds, at a speed of 90 feet per minute. These variations 
are not in themselves too serious, but should they fall in phase with 
other cyclic changes then the resulting flicker would be noticeable. 
Stocks having cyclic and random sensitivity have been submitted 
for use. In general these defects have been minimized. 

Irregular camera motion is one of the worst offenders at present, 
but the motor system is not blameless and can be the cause of flicker. 
For some time the interlock motor system, when used, was con- 



[J. S. M. P. E. 

demned for all this trouble; the fact is that the interlock motor sys- 
tem was not the source, but its basis of operation allowed the trouble 
to persist and frequently amplified it. It would appear that, to 
obtain a steady exposure, the speed of the rotating shutter, which 
exposes the film, should be as smooth and constant as the movement 
of film .through a sound recorder. A great deal of time and money has 
been spent by sound equipment manufacturers and users to reduce 
flutter, and as previously mentioned, picture flicker is nothing more 
than flutter. However, cameras generally use a slipping belt di- 
rectly coupled to the shutter shaft for a film take-up mechanism. 
Belt condition greatly influences the steadiness of take-up, and each 
instant that a sudden change in load occurs the motor system reflects 








N ~^ 

^ ^ 

. ' 

3 .5 1.0 1.5 
i. 2. Chart of section of picture negative in which photographic 
r is just perceptible, representing about 3 per cent in transmis- 

that change. Even with a motor having no resilience, changes in 
load can cause flicker. The camera undoubtedly contains mechani- 
cal inductances and capacities (which would include the shutter, 
motor rotors, gears, backlash, motor air-gap, flux, etc.) that can be- 
come resonant. Even though these reactances are inherently 
stable, the system might be thrown into oscillation by a sudden shock 
of small magnitude. This is evidenced by circumstances that have 
occurred when belt condition, mechanical looseness, and shutter 
action have all combined to become oscillatory and persistent at a 
rate well within the greatest disturbing region of Fig. 1. It was while 
working on a new motor system in conjunction with Electrical Re- 
search Products, Inc., that a full realization of the true nature of the 

Sept., 1939] 



difficulty was reached. A few clutches have been tried that gave 
varying degrees of improvement but none completely solved the 
problem. Fig. 2 shows a chart of a section of picture negative in 
which the flicker was just perceptible, representing about 3% varia- 
tion in transmission. Fig. 3 shows a similar chart having flicker 
amounting to about 8% to 10% variation in transmission. 

Those who have never seen the action of a camera shutter might 
observe the opening or closing edge of the shutter with a stroboscope, 
which is accurately synchronized with the motor, as either of these 
edges pass the aperture. Obviously, any variation in the shutter 
while it is fully open will have no deleterious effects. 


V j-* 


\A/V/ X 






FIG. 3. 

Chart showing objectionable flicker amounting to about 
8 to 10 per cent variation in transmission. 

Development of negative is suspected of causing some variations 
but no conclusive data are yet available. It should not be deduced, 
however, that the laboratory is entirely faultless. 

(B) Printing. When the printer light is supplied from a genera- 
tor or rectified alternating current, sufficient filtering must be used 
to reduce intensity changes to a small value. The ripple voltage 
should be less than 1 per cent. It is true that the normal ripple fre- 
quencies are beyond the greatest disturbing flicker rate, but the 
existence of 120 cycles in conjunction with other flicker frequencies 
produces a creeping density pattern in the projected picture. Fre- 
quencies increasingly higher than 120 cycles would undoubtedly cause 
less and less trouble. This particular effect is greatly dependent upon 
the amount of light and the density of the various parts of the scene. 



[J. S. M. P. E. 

It becomes most apparent in scenes including dark skies such as in 
night shots made with filters in the daytime. Fig. 4 shows the trans- 
mission change caused by a ripple voltage of 10 to 15 per cent. 

Periodic flicker has been traced to printer motion on an earlier 
type of machine but no data exist on machines of current manufac- 
ture. In the case under investigation the flicker was caused by the 
belt splice which created periodic film-speed changes as the stock 
passed the aperture; i. e., as the belt splice passed over the pulley the 
effective radius of the pulley was changed, causing a corresponding 
film-speed variation. Fig. 5 shows this effect under two conditions, 
that of normal operation and with an exaggerated splice. Only the 
amplitude of the variations has changed; the rate having remained 









\j \~< 

r \^- 



FIG. 4. Flicker due to printing: transmission change caused by 
a ripple voltage of 10 to 15 per cent. 

Print development, the same as in negative development, has been 
suspected of some trouble since, of two prints from the same negative, 
one may have flicker and the other not. It is true that the printer 
itself may cause this trouble but no analysis has been made. 

(C) Projection. The whole subject of background projection 
has been well covered by the work of the Research Council Process 
Projection Equipment Committee of the Academy of Motion Picture 
Arts & Sciences 1 under the chairmanship of Farciot Edouart, and a 
great number of the conclusions arrived at apply to projectors in 

The first item of particular importance in projection is steadiness 

Sept., 1939] 



of the arc lamp. This factor has been appreciated for some time and 
efforts have been directed toward its reduction. The work done by 
the above-mentioned Committee has further advanced arc lamp 

Shutter flicker in projection is still an important problem and re- 
solves into two separate factors. First, constancy of light from frame 
to frame of a particular scene; mechanical accuracy of all parts; 
lack of mechanical resonances; and a stiff or non-resilient and well 
damped motor-drive will all contribute to improvement. Shutter 
variations amounting to 7 degrees have been observed. Second, the 
effects of shutter rate and the manner of eclipsing the picture must be 
considered. The minimum rate is established by the frame fre- 





FIG. 5. 

Flicker effect in printing, due to 



quency. During the period that the shutter is open a still picture is 
being projected, but it is possible to demonstrate a reduction in flicker 
by interrupting this still picture for a short interval of time by an 
additional blade. This, of course, essentially increases the frame fre- 
quency but leads further to the possibility of other physiological 
factors. The question of two-bladed versus four or more bladed 
shutters of various dimensions, and one-sided or two-sided wipes, is 
certainly worthy of investigation. Undoubtedly the best approach to 
a real solution is by an extensive series of studies. These tests should 
be made by projecting a single frame or still picture at an intensity 
equalling that of the good theater picture and always maintaining the 
same average amount of light on the screen regardless of shutter de- 

242 L. D. GRIGNON [j. s. M. P. E. 

sign. Further, these tests should include various values of eclipsing 
times from zero up, with all the darkness occurring in one interval 
and the same total amount of darkness broken up into two or more 
intervals, single versus double wipes, and instantaneous versus 
dissolving wipes. After reaching a definite conclusion for the most 
satisfactory combination, various periodic rates of irregularity could 
be superimposed upon the shutter action to obtain more definite and 
scientific data on this particular flicker effect. 

(D) Background Projection. Background projection suffers from 
all the above ailments except one, with the additional penalty of 
having all defects increased two-fold under certain circumstances. 
The one exception is shutter flicker in the background projector, 
and this can be eliminated only by careful synchronization and by 
making either the camera shutter or projector shutter sufficiently 
greater than the other so that the irregularities of the two will not 
overlap. If the synchronism does not remain accurate overlapping 
causes a disastrous result. 

The author appreciates that all the matters discussed in this paper 
are controversial, particularly when so little concrete evidence can 
be presented; but certainly enough is on hand to indicate that 
progress along this line of endeavor should be stimulated. Much 
work remains to be done, and this work must be coordinated in such 
a manner that all persons involved in the final result on the screen 
work toward the same end. Like so many problems in complex 
art, it will not do much good for one branch to assume that its con- 
tribution is commercially sufficient. Taken by itself it might be, but 
when put in combination with other units to form the complete sys- 
tem, the final result may not be good, due to additive effects. 

The purpose of presenting this paper has been to call the industry's 
attention once again to this serious problem, to indicate present 
major sources of difficulty, and to discuss them in a limited manner; 
and also to offer whatever assistance our results may provide to those 
who are qualified and equipped to carry out further studies. 

The author wishes to acknowledge the assistance of all those 
who have contributed to this study in work, thoughts, and sugges- 
tions, and, in particular, the following: L. L. Ryder, A. F. Edouart, 
and A. Aton (Paramount Pictures, Inc.) ; A. L. Holcomb and C. R. 
Sawyer (Electrical Research Products, Inc.); and H. R. Berry 
(Lockheed Aircraft Corp.). 



1 "Recommendations on Process Projection Equipment," Research Council of 
the Academy of Motion Picture Arts & Sciences. /. Soc. Mot. Pict. Eng., XXXII 
(June, 1939), p. 589. 


MR. MORGAN: Why is it that flicker has now become such a problem? Have 
we always had flicker and not noticed it, or have we been adding small distortions 
to the photographic processes so that they now add up to make a noticeable flicker? 
How do you determine what you say is appreciable flicker? What is the per- 
centage of flicker when you begin to worry about it? 

MR. GRIGNON: We, as sound men, are interested in picture problems, first, 
because we are interested in improving our employer's product; and second, 
the Sound Department is generally connected with the Camera Department 
through the necessity of interlocking motor drives. For years we have blamed 

8-mm. M.P. Studio Trim New 8-mm. 7-mm. Trim 

FIG. 6. Light-steadiness curves. MR-40 Duarc lamp; 40 amperes, 37 arc 


the motor system for most of the causes of flicker but, as pointed out in the paper, 
we have found that it was not the motor system itself, but generally the type of 
operation, which permitted the flicker to exist and did nothing to damp it. 

The motor system was not faultless but we did need a new motor system from 
that standpoint. Flicker has always been in pictures to some extent but I can 
not give you a complete answer to that question. Perhaps one of the reasons 
we notice flicker more easily now is that we are using more brilliancy in projec- 
tion. With better pictures on the screen it is very definite that there will be an 
increase in intensity of screen flicker and that it will be more noticeable. 

In answer to the last question: the amount of perceptible flicker was deter- 
mined by having a number of persons observe results of pictures taken under 
various conditions, and then obtain the average transmission differences of the 
samples that were considered as just perceptibly degraded. 

244 L. D. GRIGNON [j. s. M. P. E. 

A discouraging thing about this flicker problem is that flicker has not existed 
day after day and week after week. Flicker has been definitely an intermittent 
problem, and during times of serious trouble has demanded the attention of many 
men, who, however, to my knowledge have not yet arrived at a true and final 
answer. The intermittent nature of this problem is undoubtedly due to the 
many factors involved. With all the work that the sound engineers have done 
on flutter it seems odd that a situation should exist that requires the same quality 
of motion but very little has been done about it, and the studios think the manu- 
facturers should seriously undertake the problem because it is rather costly to 
have to make re-takes. 

MR. JOY: This is an interesting paper. As manufacturers we have always 
worked along the line? of producing a carbon which will give a steady light. In 
fact, we have been working along the very same lines which Mr. Grignon suggests-. 
As evidence of this I refer to Fig. 6, which is taken from our paper on "Recent 
Improvements in Carbons for Motion Picture Studio Arc Lighting."* This il- 
lustrates that improvement in the carbons has resulted in a very appreciable im- 
provement in light steadiness. It should be realized also that a good lamp mecha- 
nism is necessary for the steady burning of the carbon. 

In the Technical Bulletin "Recommendations on Process Projection Equip- 
ment" of the Research Council of the Academy of Motion Picture Arts and 
Sciences, specifications and suggestions are given for burning a carbon in a 
projection system under conditions which, if followed, will go a long way toward 
eliminating any objectionable flicker. This illustrates again that besides having 
a good carbon it is necessary also to burn it properly to obtain the steady light 
desirable for either background projection or other lighting applications con- 
nected with the motion picture industry. It is evident that the work of Mr. 
Grignon and also of the Process Projection Equipment Committee of the Academy 
indicates that we are all striving toward the same common end, that is, to make 
a perfect motion picture. 

Mr. Grignon stated in his paper that a flicker of around 6 to 8 cycles per 
second in frequency was most noticeable to the eye. Was thie critical frequency 
established by observation or by some theoretical consideration? 

MR. GRIGNON: With a large series of tests we finally realized that those ir- 
regularities that were causing us the greatest amount of disturbance existed in 
the region of six and eight cycles. This statement is not founded on any actual 
measurement because to make such a measurement would require a series of 
studies and other technical data involving a great deal of work. However, it 
was quite apparent that this region presented the greatest disturbing frequencies. 

MR. LAUBE : What is the reaction in regard to the way we drive the Twentieth 
Century cameras by drive from the motor to the shutter? 

MR. GRIGNON : It has been our experience that that would probably be better 
than the current type of drive. The best way to check this point is with a strobo- 
scope which is accurately synchronized with the driving motor, preferably using 
a contractor on the motor to determine the flashing periods of the stroboscope. 
Early tests with a non-synchronized stroboscope were found to be misleading. 

MR. LAUBE: We feel that we have very good motion in the shutters on the 
Twentieth Century cameras. Stroboscopic tests are quite perfect. In back- 

*To be published in a succeeding issue. 


ground projection shots it is very desirable that each frame of projected picture 
remain on the screen for a longer time than the total length of time the camera 
requires to record it. When I refer to the length of time the projected picture 
remains on the screen, I am not including the element of time during which the 
shutter in the projector is uncovering or covering the aperture, but only the time 
when the picture has its full value on the screen and is not being dissolved in or 
out by the projector shutter. If this time period is long enough to overlap that 
of the camera's total recording time period, I feel that the condition thus de- 
scribed would be most ideal for flicker elimination in background projection shots. 

MR. GRIGNON: In background projection work that is important. If we 
assume a seven-degree variation in shutter operation, which we have observed, 
then, the projection shutter should be fourteen degrees wider than the camera 
shutter, or vice versa. In using a three-head or three-projector type for projec- 
tion there is some improvement because the change in any one shutter affects 
only one-third of the total light and the result is only one-third as great also, there 
being three shutters, the change is more at random and the defect is not as serious. 

MR. KELLOGG: Would you consider a disturbance that might occur every four 
seconds as disturbing? 

MR. GRIGNON: Offhand I would say that such a disturbance, unless of large 
magnitude or occurring simultaneously with other factors, would not be dis- 

MR. RICHARDSON: In the illumination of motion pictures, we do not have any 
rotary arc that carries the rotation of the positive carbon as high as 15 rpm. 
Practically all the modern lamps of the high-intensity rotary type operate at a 
positive rotation speed from 6 to 12 rpm. 

There is another potential cause of flicker in the taking of pictures. In some 
studios it has been a practice to stop the rotation of the positive carbons during 
picture takes. This has come at the insistence of the sound recording departments 
in an attempt to reduce the mechanical noises from the high-intensity spot equip- 
ment. Some time ago a Committee of the Academy of Motion Picture Arts and 
Sciences made a study of arc noise reduction. For the test work done in this 
connection, we had available to the Committee one of the quietest stages in the 
industry, a stage on which the ventilation system was made inoperative and the 
ground-noise cut to a very low level. The test was made with a battery of ten 
150-ampere high-intensity arc spots centered around a microphone of the type 
used for recording dialog, in a semicircle having a 25-ft. radius. Studies were 
made to ascertain the effects resulting from bringing the arcs into good trim and 
then cutting the motors off. Our interest was primarily in sound. Careful 
records of this test were made, and they are available through the Academy to 
those who wish to study them. 

While it is unquestionably desirable to eliminate all possible noise in these 
arcs, the test revealed that the mechanical noise is a small factor of the total 
noise, the principal disturbance coming from the phenomena in the electrical 

The point I want to bring to your attention particularly is the decay of light 
and the production of flicker in the photographing illumination. When motors 
are cut off on the studio arcs, not only is the rotation stopped, but also the feed 
of the positive and negative carbons. Under these conditions the arc rapidly 

246 L. D. GRIGNON Q. s. M. P. E. 

becomes unsteady. The decrease in the illumination is practically linear, and 
five minutes of operation after the motors have been cut off produces a decrease 
of over 50 per cent in the total illumination. The unsteadiness of the illumina- 
tion during this period of decrease steadily becomes worse and would surely, 
at the end of three minutes, be highly contributive to flicker. However, it 
might be well to note that the flicker in illuminating sources under these condi- 
tions are random in each lamp. Those who are familiar with the studies made 
by the Sound Reduction Committee will agree, I believe, concerning the in- 
advisability of shutting off the motors on rotary arcs during photographic opera- 
tions, both from the standpoint of its increasing light flicker and actually in- 
creasing, rather than decreasing, the noise produced by the arc's operating in a 
very erratic manner. 

MR. GRIGNON: From what Mr. Richardson says the arc rotational speed is 
about one revolution every four seconds and, as stated before, that in itself, as a 
result ot our studies, would not show in the projected picture. However, in 
split-screen work if the lamp intensity changes should happen to be in opposition 
on the two halves of the picture then a push-pull effect obtains which is definitely 
disagreeable to the observer. 

MR. RICHARDSON: Many studies and tests have been made relative to these 
light-projection problems. Commercial rotary arcs are in operation today that 
limit the light fluctuation from rotation of the positive carbon to a variation of 
==3 per cent, and it is possible by refinement to reduce the effect still further. 

MR. GRIGNON: With respect to the rest of your point, Mr. Richardson, I 
do not have to defend the sound departments with regard to stopping the arc 
motors. I think that the motor noises that existed were of a periodic type that 
attracted attention. You are perfectly correct in saying that after a short 
period of time the general arc noise is increased. I also heard the tests you speak 
of. The noise, after a definite length of time is increased, but, however, that 
noise can be more readily tolerated since it is a random type of noise which 
shows up more as a constant sort of background behind the scene. Further, the 
tests of which you speak indicate clearly that even with lamps having acoustic 
treatment the motor noise is objectionable and for a period of about three to four 
minutes after disconnecting the feed motors a definite improvement in noise is 
obtained. Incidentally, three to four minutes time represents 270 to 360-foot 
takes, which are generally above the average take length. 

MR. RICHARDSON: In photographing Shirley Temple's first Technicolor 
picture at 20th Century-Fox, the sound engineers had a particularly difficult 
situation. In this picture there were a number of very intimate scenes which 
required the recording of the children's voices talking either together or to their 
nurse. The intimacy of the scenes and the diminutive voices had to be recorded 
against a background of almost constant level. When some of the first scenes 
were taken we were called upon to analyze a problem of underexposure which 
Technicolor encountered in their photographic operations. This led to a special 
study of the decrease of light resulting from shutting off the motors of the rotary 
arcs, which had been the practice in taking these intimate sequences, 
records of these studies are available to anyone who is particularly interested 
studying this effect. 

Some of these studies have revealed the effect of the "flash" type of flicker. 


which I think very definitely falls within this subject. These "flash" flickers are 
particularly prevalent in high-intensity arcs when the carbons are not operated 
at their normal consumption rates, and when the arc craters become unsym- 
metrical, which is always the case when the positive carbon which has been pur- 
posely designed to rotate, has its rotation slowed down, or is stopped. 

MR. CRABTREE: How do you measure the flicker? 

MR. GRIGNON: We are not equipped to make accurate determinations of the 
various flicker effects. The pictures used for observation were made by photo- 
graphing a neutral gray wall under various conditions of the mechanism, take- 
up belt, motor system, shutters, etc., and by actually applying periodic disturb- 
ances to the motor shaft. These pictures were then submitted to the various 
observers for their comments. Later the test-films that were of interest were 
measured throughout their length by a method employing what was essentially a 
recording densitometer having a relatively slow time response so that the density 
of each frame was somewhat averaged. Strictly comparative results were ob- 
tained by this method and the final answer obtained was, as noted in the paper, 
3 per cent difference in transmission for perceptible flicker. More accurate 
methods of obtaining the data would certainly be of value and would definitely 
be required in order to separate the various types of flicker. 







Summary. A brief account is given of the development of the process projection 
art, and the difficulties that arose due to its intial haphazard evolution. 

Following this is an account of the work of the Process Projection Equipment 
Committee of the Research Council of the Academy of Motion Picture Arts and 
Sciences, leading up to the recommendations of the Committee with regard to the 
projection equipment used in the process. 

The motion picture industry's utilization of the projection back- 
ground or transparency process of composite cinematography has 
become more and more important every year. Scarcely a picture is 
made which does not in some degree utilize this process; some have 
employed it to such an extent that as much as 50 to 60 per cent of the 
production's released footage has been produced on the process stage. 

When considering this extensive use of process shots, it should be 
kept in mind that while the process naturally lends itself to the pro- 
duction of out-and-out "trick shots," its routine application is rarely 
in this category. The transparency process is not used as a means of 
fooling the public, but simply as the best and often the only method 
of securing scenes which would be too difficult, too costly, or too 
dangerous to film by conventional methods. The expense of sending 
a large company of stellar players, "extras", and technicians to do 
extensive filming on a distant location would under modern condi- 
tions reach a staggering total. But thanks to the present efficiency of 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received June 
26, 1939. 

** Transparency Department, Paramount Pictures, Inc., Hollywood, Calif. 


projection background cinematography such scenes can now be 
photographed in the studio with equal effectiveness, and at a con- 
siderably lower cost. 

It can be readily seen, therefore, that the application and advance- 
ment of this process have assumed not merely a technological im- 
portance, but an economic importance vital to the continued financial 
success of the motion picture industry. 

The process developed spontaneously as technicians throughout 
the industry discovered methods and materials which made it 
practicable. These key developments were the introduction of sound, 
which brought the means of electrically synchronizing the composite 
or foreground camera with the background projector, and the intro- 
duction of the first supersensitive panchromatic emulsions, which for 
the first time afforded the high film sensitivity necessary for re- 
photographing the projection background image. With these ele- 
ments available, it was inevitable that cinematographers in practi- 
cally every major studio should put them together to form in actuality 
a system which for years many of us had pondered in theory. 

From a simple, utilitarian standpoint, this spontaneous evolution 
was something of an advantage, since it quickly gave the entire 
industry a valuable new tool, with a minimum of dissention over 
proprietary patent formalities. 

But from an engineering and industrial planning standpoint, it was 
highly unfortunate, because this haphazard evolution prevented any 
early standardization of essential methods and equipment. 

As it was, each studio's process staff built or assembled their own 
equipment, often with little or no knowledge of what was being done 
along similar lines by other studios. Inevitably a great deal of waste- 
ful duplication of effort ensued. Private manufacturers, when called 
upon to build process equipment for one studio, could seldom plan on 
selling similar units to other studios, not because of any patent or 
other restrictions, but because such units in all probability would not 
coordinate with the other studios' individual systems. 

As a result the industry paid the penalty of using custom-made 
equipment. This was evident not only in the inevitably high cost of 
equipment, but in the lack of availability of items such as especially 
designed super-speed lenses and steady, silent, ultra-powered lamp 
houses, the design and construction of which would entail such ex- 
tensive research that no major manufacturer could afford to under- 

250 A. F. EDOUART [j. s. M. P. E. 

take the project knowing that his total sales would be restricted to but 
one or two units. 

Yet the technical problems underlying the design of process equip- 
ment were crying for skilled engineering and coordinated research. 
Since the very inception of the transparency projection process, it had 
been found that ordinarily available projection equipment for this 
type of work is principally composed of an assembly of units never 
originally designed or engineered to be combined and worked to- 
gether in such a capacity. Basic elements of these assemblages were 
never intended to meet such strict and exacting requirements as have 
been imposed by the consistent demand for higher-quality rear pro- 
jection results, and by the ever-increasing physical scope demanded 
by the increasing need for the process. We must not merely project 
our background images on a translucent screen and rephotograph 
them without subjects in the foreground: we must do it with such 
mechanical and photographic precision that both the foreground 
subject and the background projected picture are of such quality as 
to appear as if they were photographed at one and the same time. 
The increasing use of the process demands increasingly great physical 
scope in other words, the ability to use background screens of in- 
creasingly greater size. 

These requirements, in a word, mean Maximum Light Delivery 
with the following primary requisites: Absolute Steadiness of the 
projected picture with a Minimum of Light Variation on the screen 
and Increased Efficiency of the Light and the Optics which Transmit It. 

Over a year ago the Research Council of the Academy of Motion 
Picture Arts and Sciences realized that it was necessary to take steps 
to correct this harmful situation, and to coordinate all the various 
ideas of the different studio Process Departments so that manufactur- 
ing companies would not be trying to manufacture a separate type 
of equipment for each producing studio. To this end, in March, 
1938, the Research Council's Process Projection Equipment Com- 
mittee was appointed. On February 2, 1939, the Committee's Re- 
port on Recommendations on Process Projection Equipment was 
approved by the Research Council for publication and distribution 
throughout the industry. 

In the preparation of this Report, eleven meetings and two demon- 
strations, consuming approximately one thousand man-hours, were 
held, and at least an equal amount of time was consumed by the Com- 
mittee Chairman and members in conferences, in preparing for meet- 


ings, tests, and demonstrations, and in preparing the Report itself. 
The Report therefore represents over two thousand man-hours of 
technical effort and combines the views of approximately fifty ex- 
! perts on the field of process projection. 

It is only fitting to state here that the Research Council and the 
Committee are greatly indebted to all the many manufacturing firms 
consulted for their participation in this program. We are particu- 
larly appreciative of the cooperation of the National Carbon Com- 
pany for sending its Development and Control Laboratory Director, 
Mr. David B. Joy, to Hollywood in connection with the development 
of carbons for process projection use, and to the Bausch & Lomb 
Optical Company for sending its representatives, Mr. Haller Belt 
and Mr. Alan A. Cook to Hollywood in connection with the develop- 
ment and standardization of optical systems for process projection 
work. All these men remained in Hollywood for several weeks, con- 
ferring with the Committee and its members. 

The cooperation of such other firms as the International Projector 
Corporation, the Mitchell Camera Corporation, the Technicolor 
Motion Picture Corporation, the General Electric Company, the 
Mole-Richardson Company, Paramount Studios, RKO-Radio Stu- 
dios, and Selznick-International Studios in the work of this Com- 
mittee was also cheerfully given to an extent far greater than is 
ordinarily required of participants in the Council's program. 

It is to be regretted that at this meeting we do not have the time 
to read the name of each member of the Committee and thus to ex- 
tend the credit that is due. On this Committee there are 37 active 
members, not only cinematographers and technicians from all the 
studios participating in the Research Council program, but also ex- 
perts from the companies which are the leading manufacturers of all 
the many types of equipment used in process projection work. Their 
interest and their whole-hearted cooperation made possible the suc- 
cess of this, the first step in the Committee's program. 

We believe that the Report presents for the first time the coordi- 
nated viewpoint of the majority of Hollywood studios and their ex- 
perts on this subject. As such, it should be of value to all the studios 
and to all the manufacturers of process projection equipment. The 
text of the report was published in the June, 1939, issue of the 
JOURNAL and should be of interest to every member of the Society. 

In studying the report it should be observed that we have stressed 
the importance of faster lenses, perfect illumination, perfect registra- 

252 A. F. EDOUART [j. s. M. P. E. 

tion, and increased light. Progress in these factors is the key to prog- 
ress in the transparency projection process as a whole, and as these 
have been improved, the scope and utility of process projection has 
advanced proportionately. 

Our great underlying problem is the fact that in our composite 
shots we are dealing with two photographic irreconcilables : for the 
image of our foreground action is an original, while that of the back- 
ground is necessarily a "dupe," being the rephotographed image of a 
positive print. It is our task to combine these so perfectly as to color 
values and perspective that both appear as if they were photographed 
at the same time and place, and so that both maintain a photographic 
quality comparable to that of the production scenes (necessarily 
originals) with which they are ultimately intercut. 

As the work has progressed, and the usefulness of the process has 
increased, all of us have striven to increase the general physical scope 
of the process and to provide a means of increasing the size of back- 
projection screens we could use. When the process was first intro- 
duced, screens six or seven feet in width were about the maximum 
possible. Two years ago the largest screens that were being used by 
the various studios ranged between fourteen and sixteen feet in 
width. Last year, when we began the transparency sequences for 
Spawn of the North, we commenced with a twenty-four-foot screen; 
we ended the picture using a thirty-six-foot screen, and filling it with 
ample illumination . I am sure all will agree that this represents progress. 

As regards quality we and by we I do not by any means restrict 
my meaning to my own Department at Paramount, but I refer to 
the process experts of all the major Hollywood studios have very 
consistently succeeded in turning out process shots so convincingly 
natural that in the majority of cases the layman is not aware that the 
scene was made on the process stage. In some instances we have suc- 
ceeded in producing pictures using really large screens where it is im- 
possible even for many technicians to discern the difference photo- 
graphically between the quality of the foreground original and the 
background dupe, or to point definitely to a scene and say, "This is 
a process shot." In doing this it will be appreciated that any flicker, 
any fluctuations, or any unsteadiness in the projected background 
naturally dispels the illusion of reality and reveals the scene as a 
process shot, lessening its dramatic value. Our aim is to eliminate 
these flaws, and to extend the scope and usefulness of the process. 

The recommendations contained in the Research Council Process 
Projection Equipment Committee's Report will, we believe, further 


these aims, to the benefit of the industry and of the engineers and 
manufacturers who supply its equipment. Although these recom- 
mendations have been published but two months, they are already 
beginning to bear fruit in a concrete and satisfying manner. 

The new lamp houses required are being developed. New lenses 
of two higher speeds are being calculated. There was a very im- 
mediate response by the Bausch & Lomb Optical Company when the 
various studios signified their intention of purchasing seventy-six of 
the new Super-Cinephors. That, by the way, was within a week or 
two after the recommendations were released at the studios. The 
Mitchell Camera Company is working on a new projection head, de- 
signed in accordance with these specifications. The Mole-Richardson 
Company has made notable progress in engineering a new lamp house 
and a new grid to meet these specifications and has the first equip- 
ment practically completed. With these developments, we can anti- 
cipate notable progress in the art of process projection cinemato- 
graphy in the near future. The ability to use background screens 
fifty feet wide seems well within the realm of practical possibility. 
So, too,* do worth-while advances in steadiness, both of picture and 
of screen illumination, in picture-quality and in operating precision. 

In closing, I can think of no better way to summarize than to quote 
from the preface written for the Report by Major Nathan Levinson, 
who in addition to being Executive Vice- President of the S.M.P.E. 
is Acting Chairman of the Research Council : 

"Process projection methods continue to become increasingly im- 
portant : Economically, they offer opportunities for still greater sav- 
ings in production costs. Technically, developments in equipment 
and technic continue to expand the possibilities in this field until some 
day it will be the exception, rather than the rule, to send a cast on a 
distant location. 

"Artistically, as the equipment and technic are further developed, 
the extent of their use will be limited only by the imagination of the 
production personnel; whereas up to the present time, the equipment 
has been the limiting factor, and only the ingenuity and resourceful- 
ness of the technicians have made its wide use possible." 

To these ends, the Research Council's Process Projection Equip- 
ment Committee is continuing its work. We consider the present 
recommendations a vital initial step in that work, but further co- 
ordinated activity will build upon that foundation. For the present, 
we hope that the industry will accept these recommendations and 
profit therefrom, 


Summary. A microphone is described which has uniform directivity over a ; 
wide frequency range. This is made possible by placing in a single instrument a ' 
dynamic type pressure microphone element and a ribbon type "velocity" element, 
and electrically equalizing the outputs before combination. The resultant directional ; 
pattern is a heart-shaped curve or cardioid, giving a fairly wide pick-up zone in front ; 
and a substantial dead zone at the back of the instrument. Because of the unusually \ 
rugged ribbon employed, the new microphone is much less susceptible to wind noise 
than ordinary ribbon types. Housed in an aluminum case, the microphone weighs ', 
only 3 l / t Ibs. High output level, low impedance, and high quality, together with the \ 
excellent directivity, promise to make the cardioid microphone an important tool for \ 
the motion picture sound engineer. 

A microphone, in sound motion picture systems, bears the same 
relationship to sound that a camera does to light. The special j 
lighting technics of motion picture photography have been so widely ] 
publicized that everyone is aware of the steps taken to provide a \ 
visual illusion. It is also well known 1 that similar technics are re- 
quired in sound recording to create the illusion of presence of the I 
sound with the screen action. The balancing of these two arts, only 
just now being introduced to broadcasting through television, has 
long been practiced in the motion picture business; with the result 
that a particular set of microphone requirements have been evolved 
which are peculiar to the picture industry. 

There are certain factors to be considered at the microphone which 
can not be handled at any other spot in the system. It is naturally 
expected that the microphone shall have a smooth frequency re- 
sponse over a wide frequency range, that the nonlinear distortion 
shall be at a minimum, and that the phase relationship among the 
frequency components of a complex sound shall be kept within 
bounds. Moreover, not only must the microphone be adaptable 
to use under a variety of acoustic conditions, but all the microphones 

*Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received May 
7, 1939. 

**Bell Telephone Laboratories, New York, N. Y. 


must produce records which can be intercut so that a continuity of 
sound quality is preserved throughout the performance. 

Still another requirement, and a severe one, is imposed on the mi- 
crophone by the type of sound system employed. Most sound sys- 
tems in use today are monaural, which means that all the transmitted 
sounds come from one source, the loud speaker. In normal listening 
we are often interested in sounds coming from a particular direction, 
and our aural senses have the peculiar property of being able to focus 
our attention on such sounds. In monaural listening, however, we 
can not use these focusing powers to separate unwanted sounds which 
are transmitted. Therefore these unwanted sounds must be elimin- 
ated at the original location, and this may be done in two ways, either 
by controlling the sources of these sounds 2 or by making the micro- 
phone directional. 

Aside from extraneous noises such as camera click, the principal 
class of sounds it is desired to exclude consists of reflections which 
render the sound reverberant. Now a certain amount of reverberant 
sound is necessary to achieve naturalness since we are accustomed to 
listening in rooms or auditoriums, but in monaural listening, where 
we lose the focusing powers of our aural senses, reverberation becomes 
much more noticeable or unnatural. If the reflected sound can be 
reduced either at the reflecting surfaces or at the microphone, then 
the reverberant effect may be reduced to the point where the re- 
produced sound will be natural. Moreover, in sound motion picture 
systems, control of this reverberant effect helps provide fore and aft 
presence in the reproduced sound. 

The microphone to be described in this paper enables, for the 
first time, many of these pick-up conditions to be controlled at the 
microphone itself. This is made possible by providing the choice of 
three different directional patterns, so that the instrument is truly 
a "mechanical ear" which can help perform the functions of our nor- 
mal aural senses in centering our attention on direct sounds. There 
still remains unexplored, however, the practical applications of this 
microphone to the operating technics of sound motion pictures. 

Directivity in a microphone can be of the greatest service only if it 
is independent of frequency and only if the discrimination in the 
undesired directions is high. The failure of most commercial direc- 
tional microphones in the past to meet one or both of these require- 
ments accounts for the fact that the field is virtually unexplored. It 
is the purpose of this paper to describe a microphone which has uni- 



form directivity over a wide frequency range, responds to sounds 
corning from the front but is dead to those coming from the back, and 
yet meets many of the usual specifications of the sound motion picture 
studios. This new microphone, known as the Western Electric 
639- A, and shown in Fig. 1, combines in to one instrument the popular 
features of both the dynamic pressure and ribbon velocity types. It 
is well known that by virtue of their fundamentally different princi- 
ples of operation these two elements yield in proper combination a j 

heart-shaped directional pattern 
known as a cardioid. 3 

s/tf^f*"~ ^ e principle of operation of 

/[iff / the dynamic or moving coil ele- 

f[\ 1 1 ment is illustrated in Fig. 2. It 

consists essentially of a movable 
membrane or diaphragm closed 
off on one side by a hollow case, 
which should have small impe- 
dance compared to that of the 
diaphragm itself. Excess sound j 
pressure at the diaphragm surface 
will cause it to move inward 
against the normal air pressure 
within the housing. Because 
sound can flow around corners, 
this type will respond to sound 
arriving from any direction, and 
excess pressure will always cause 
the diaphragm to move inward 
regardless of the angle of inci- 
dence. This accounts for the label 
"nondirectional pressure ele- j 

ment." A coil attached to the diaphragm and placed in a magnetic 
field transforms the mechanical vibrations into the desired electrical 
oscillations. Because the diaphragm and coil assembly permits an j 
efficient magnetic structure and is of high enough impedance to j 
permit the use of a small hollow case, this moving-coil type can be j 
made very small without sacrificing sensitivity a decided advantage 
over many other types of pressure elements. 

The ribbon unit works on the quite different property of a sound- 
wave, namely, the variation of pressure with respect to distance; 

FIG. 1. Western Electric 639- A 
cardioid directional microphone. 



the rate of variation being known as the pressure gradient. Thus, if 
a ribbon is placed in a sound field as shown in Fig. 3, there will exist 
a pressure difference between the front and back of the ribbon by 
virtue of the acoustic path distance L. The ribbon will naturally 







FIG. 2. (Upper.) Principle of dynamic-pressure 
microphone. Neutral pressure in case can change only 
by motion of diaphragm. Excess pressure on outside 
will push diaphragm in regardless of direction of sound 

FIG. 3. (Lower.) Principal of ribbon type pressure 
gradient microphone. 

tend to move in the direction of diminishing pressure. If the ribbon 
is hung in a magnetic field, its motion will cause an electrical voltage 
to appear across its length. Observe carefully, for this is the key to 
the problem, that the motion of the ribbon will reverse with reversal 


R. N. MARSHALL AND W. R. HARRY [j. s. M. P. E. 

of the direction of the incident sound-wave. It can be readily seen 
that if the wave approaches the ribbon on edge, the pressure will be 
equal on both sides; the motion will be nil; and the response zero. 
Actually, the sensitivity is proportional to cos 6, the angle of sound 
incidence, as shown in the two-looped characteristic Fig. 4 (B). The 
device is "two-faced" or bi-directional since it picks up sound coming 
from front or back but not from the sides. 

The motivating force on the ribbon is in phase with the pressure 


FIG. 4. A Ideal characteristic of a non-directional 
pressure microphone. 

B Ideal characteristic of a bi-directional pressure 
gradient microphone. 

C Cardioid directional characteristic resulting from 
combination of A and B. 

gradient which is 90 degrees out of phase with the pressure, and in- 
creases nearly linearly with respect to frequency. This is offset 
over the operating frequency range by keeping the ribbon impedance 
a mass reactance which also increases with frequency and introduces 
a 90-degree phase-shift. As a result, the motion of the ribbon is 
actually in phase with the particle velocity of the air which accounts 
for the term "velocity" microphone. A "velocity" type is merely a 
special case of the pressure gradient variety. 



The ribbon pressure gradient element may be thought of as pro- 
viding the directional ingredient, but in a "two-faced" form which 
has severe limitations of use. The non-directional dynamic element 
is added to annul one unwanted loop of the ribbon response and to 
transmute the directional pattern into the more desirable cardioid 
type. This is possible, because, as explained above, the ribbon 
motion reverses with reversal of sound direction so that the unchang- 
ing dynamic only annuls one loop and augments the other (Fig. 5). 
If the maximum sensitivity of both types is represented by the quan- 
tity 1, then the output of the series combination will be given by 1 + 
cos 6 which in polar coordinates has the shape shown in Fig. 4(C). 
Since the word uni-directional implies one direction only, the name 
"cardioid directional microphone" has been adopted here as more 















FIG. 5. Principle of the 639-A cardioid directional microphone. 

descriptive of the combination of a pressure gradient and pressure 

Although this explanation serves to present the general idea under- 
lying the new microphone, the complete analysis is quite compli- 
cated; principally because neither element follows the simple as- 
sumptions made above. In the first place, the physical sizes of the 
elements, for the sake of sufficient output level, are so large that the 
assumption that the dimensions are small compared with the wave- 
length is not met. In the second place, the magnitude and phase of 
the outputs are not the same over the frequency range so that can- 
cellation of one response loop of the ribbon type can not be achieved 
in simple fashion. The solutions of these problems are so interrelated 
and tangled together that they can not be considered separately in 



rigorous fashion, and space prevents the complete unravelling of the 
story. After all, it is the results that we are principally interested 
in so we will only glance at the highlights of the design problem. 

The dynamic element is the same compact unit used in the Western 
Electric 630- A 4 nondirectional microphone. The problem of getting 
the diaphragm of the dynamic unit in close proximity to the ribbon so 
as to minimize phase differences caused by the distance of separation, 
and at the same time prevent serious disturbances of the normal opera- 
tion of each element due to the 
presence of the other, is solved by 
the electromechanical structure 
shown in Fig. 6. A special ribbon 
magnet structure has been devel- 
oped using a highly effective per- 
manent-magnet steel, which per- 
mits a fairly open arrangement. 
The housing of the dynamic unit 
has been reshaped and streamlined 
so that its presence does not seri- 
ously affect the operation of the rib- 
bon. In addition this housing 
encloses the ribbon transformer, 
electrical equalizer, and switch (see 
Fig. 6); supports the ribbon wind 
screen housing; and provides the 
mounting and terminal facilities in 
the form of a projecting cylindrical 
plug. It may be noted that the 
size of the device is dictated largely 
by the ribbon element which, be- 
cause of the wide air-gap to ac- 
commodate the ribbon, requires heavy permanent magnets. More- 
over, the ribbon is susceptible to air currents and must be sur- 
rounded by a cloth screen a few inches away, which reduces the 
flow of air through the instrument and yet has a negligible acoustical 

Since the screen, by requirement, has no effect on sound transmis- 
sion, its shape is not quite as functional as in the case of the dynamic 
housing where the hard shell forces sound to flow around it. Conse- 
quently the screen was styled in such a manner as to convey as much 

FIG. 6. Cross-section of the 
639-A cardioid directional micro- 


as possible the idea of directionality. For this purpose an aero- 
dynamic motive has been employed in which the bulbous end is the 
front. At the same time, however, every inch of space has been 
utilized so that the overall size is the minimum consistent with oper- 
ating requirements. The housing is fashioned of a cast aluminum 
alloy, and is finished in aluminum grey with the horizontal lines in 
polished metal. The overall height of the microphone, including the 

FIG. 7. For stage work where large angles of 
inclination are required, the universal 11- A at- 
tachment, a combination suspension mounting 
and swivel holds the microphone at the center 
of gravity and friction joints permit quick set- 
ting of the instrument in any desired direction. 

plug terminal mounting, is 7 1 /z inches; and the weight is approxi- 
mately 3 1 /4 pounds. 

The 639-A is designed to mount directly on a floor-stand as shown 
in Fig. 1 . No tilting of the microphone is normally necessary since, 
as will be explained later, because of its broad pick-up angle, sufficient 
adjustment may be obtained by setting the stand in the right direc- 
tion. For stage or set use where hanging or other placement requires 
a tilting feature, a universal swivel mounting is available. This 


R. N. MARSHALL AND W. R. HARRY [j. s. M. P. E. 

mounting, shown in Fig. 7, suspends the microphone from its center 
of gravity and friction joints allow setting of the instrument in any 
direction. The mounting may either be suspended by cords or at- 
tached to a floor or desk stand. 



DB 20 LOG 


11.4 D p 











Z -45 
% -90 


<r -225 







: = 





















50 70 100 200 400 6OO 1000 2000 


4000 6000 10000 

FIG. 8. (Upper.) Effect of difference in phase and 
magnitude on the resultant of the sum and difference of 
two vectors. This illustrates the necessity of carefully 
matching the outputs of the two microphone elements in 
order to obtain good discrimination. 

FIG. 9. (Lower.) Relative phase-angle between out- 
puts of dynamic-pressure element and ribbon-pressure 
gradient element. 

Another feature of the new microphone is the three-way switch 
located at the lower rear of the housing. This allows the choice of 
the dynamic or ribbon units individually as well as the combination 
which gives the cardioid directional characteristic. Essentially then, 


the owner of a 639- A has three microphones in one, non-directional per- 
formance from the dynamic, bi-directional from the ribbon, and car- 
dioid-directional from the two together. The responses of the bi- 
directional and non-directional elements, however, have been adjusted 
to permit the best results in the cardioid combination, and therefore 
may require slight equalization, at high frequencies, if takes, using the 
three different types of performance, are to be intercut. 

With the physical size and construction of the pressure and pressure 
gradient elements set, there still remains the question of whether 
their electrical outputs are sufficiently alike in magnitude and phase. 
To understand the significance of this problem, let us examine for a 
moment the diagrams of Fig. 8, which illustrate, the effect of relative 
differences in the output voltages. Note that whereas the addition 
resultant of the two vectors is not very critical, the subtraction re- 
sultant is very sensitive to variations. A difference in magnitude of 
3 db and phase of 30 degrees permits a discrimination between front 
and back of only 10 db. Complete cancellation is evidently a very 
difficult thing to achieve. 

For the present case Fig. 9 represents the relative phase between 
the pressure and pressure gradient elements for sound incidence of 
180 degrees which is the direction for which we desire complete can- 
cellation. It is obvious from Fig. 9 then, that aside from magnitude 
differences, the phase characteristics will permit good results only in 
the middle frequency range, and some corrective steps are necessary. 
Now it so happens that most of the phase differences are attributable 
to the pressure element, and are difficult to avoid unless we sacrifice 
ruggedness and size. The question naturally arises as to whether 
this phase shift has a serious bearing on the quality of the transmitted 
sound and should be avoided. Fortunately, this question has been 
studied in some detail by research physicists, and it has been found 
that the human ear has difficulty in distinguishing between a system 
with considerable phase distortion and one without. 5 In other words 
wave-form is of little significance physiologically, and it may be as- 
sumed that other factors are considerably more important from a 
quality standpoint. Therefore, in the present problem, emphasis has 
been placed on reducing the relative phase-angle to a minimum with- 
out reference to the absolute phase. Moreover the problem natu- 
rally divides itself into two parts, high frequency and low frequency 

From Fig. 9 it is apparent that considerable equalization is neces- 


R. N. MARSHALL AND W. R. HARRY [j. s. M. P. E. 

sary to bring the phase exactly in line, and means for doing this are 
rather complicated. Consider for a moment the response characteris- 
tics for a dynamic type microphone alone given in Fig. 10, and note 
that at high frequencies this type is directional already. Hence, as a 








, , - 









A 20 " 


o , 

-> 1C 



1 ( 










II ' 


30 5O 70 1OO 200 400 6OO XXX) 2000 4OOO 6OOO OOOO 



DB = 1 












: a 

















30 50 70 100 2OO 4OO 600 1000 2000 4000 6000 10000 


FIG. 10. ( Upper.) Field response of a dynamic-type 
microphone for several angles of sound incidence. 

FIG. 11. (Lower.) Field responses of ribbon and 
dynamic elements, illustrating the manner of equalizing 
and matching the outputs up to 3000 cycles, where the 
ribbon element is then filtered out. 

solution to the problem, the ribbon element is used to provide di- 
rectivity at the lower frequencies and is filtered out at high frequencies 
leaving the dynamic element do the work alone. The filter network 
chosen acts also as a phase equalizer, and Fig. 11 shows just how the 
transition is accomplished. Curve A is the 180-degree response of 


the dynamic element and is the curve to be matched by the ribbon 
since it is for 180 degrees that we desire complete cancellation. Curve 
B is the response of the ribbon and has been made of such a shape 
that the loss in the filter matches the two outputs very closely up to 
about 3000 cycles, and then spreads them rather sharply, as shown by 
curve C. The relative phase between the dynamic element and the 
ribbon element with the corrective network is given by curve B of 
Fig. 9. 

The results are now clear. Good cancellation may be expected 
up to 3000 cycles since the responses agree closely both in magnitude 
and phase. From 3000 to 8000 the ribbon element stays in phase but 
drops off in magnitude so that the directivity resulting from cancella- 
tion gradually diminishes with increasing frequency, but this has been 
made to coincide with the increasing directivity of the dynamic ele- 
ment. Consequently, there is no loss of directivity in the transition 
region and at 8000 cycles where control of the ribbon phase is lost, it is 
contributing so little to the output that it doesn't matter. Still 
another bit of matching occurs in the zero incidence response where 
the falling off of the equalized ribbon response is offset by the rising 
characteristic of the moving coil unit. The actual performance of 
the microphone is shown by the response curves of Fig. 12. 

The problem of equalizing the low end of the two elements to 
achieve the discrimination shown in Fig. 12 is so closely associated 
with that of reducing wind noise caused by fluttering of the ribbon 
that the two may be discussed simultaneously. From the standpoint 
of wind noise, the thicker the ribbon material, the stiffer the ribbon, 
and the greater the stability. Offhand, since the sensitivity is pro- 
portional to the mass of the ribbon, it would seem that a thicker rib- 
bon would result in too high a loss but this is partly offset by the 
reduction in electrical resistance. A compromise was selected in the 
form of a ribbon 10 to 15 times as stiff as those normally employed 
in ribbon microphones, and yet causing a loss of only 1 db. Such a 
ribbon, however, resonates at approximately 45 cycles causing a 
large peak in the low end response, unless provision is made to sup- 
press it. Placing a shunt inductance and resistance, however, 
across the ribbon terminals not only introduces damping of the 
ribbon through electromagnetic coupling, but also shifts the phase 
to correspond to that of the moving coil element. This is illustrated 
by curves C and D of Fig. 9 and D and E of Fig. 11. 

A thick ribbon formed with corrugations over its whole length, a 


R. N. MARSHALL AND W. R. HARRY [j. s. M. P. E. 

form commonly used for thinner ribbons, has harmonic modes of 
vibration within the useful frequency range. This problem is solved 
by a unique ribbon form. The ribbon is given a cylindrical curvature 
over most of its length and is corrugated at each end so that the ac- 
tion is more like a bar hinged at each end. By this means the funda- 
mental mode of vibration is affected very little but other modes are 
effectively suppressed. The curves of Fig. 13 make clear the extent 

30 30 10 100 200 400 600 1000 20OO 400O 6000 100OO 



S' 85 

30 50 70 100 200 400 


Field response of a representative 
cardioid directional microphone. 
Low-frequency response of rib- 

FIG. 12. (Center^) 
model of the 639- A 

FIG. 13. (Lower.) 
bon elements. 

A new type thick ribbon with longitudinal stiffen- 

B Thick ribbon with lateral corrugations. 

to which the microphone response is smoothed out by the action of 
this new type ribbon. 

Besides achieving a smooth response that matches that of the 
dynamic both in magnitude and phase, the stiff ribbon reduces wind 
noise to a level approximately 10 db lower than that encountered with 
usual ribbon microphones. This is of considerable importance since 
it permits the microphone to be used more freely outdoors where 
breezes are often unavoidable, and for the first time makes "panning" 



of a ribbon type from a boom practicable. Also there is less likeli- 
hood of damaging the ribbon by exposure to a sudden gust of wind, 
especially since mechanical stops are provided which prevent 
motion of the ribbon beyond the elastic limit of the material. This 
does not mean that the ruggedness of the dynamic type has been 
matched, but it does assure a wider field of application for a ribbon 
element than has been possible heretofore. 

The network required to accomplish these results is quite simple 
electrically, but the physical size of the condenser required is quite 
inconvenient if the circuit is to be included in the microphone. 
An obvious way around this would be to transform the condenser cir- 
cuit up to a higher impedance where a smaller capacity could be used 

FIG. 14. A special 3-winding transformer makes pos- 
sible the reduction, illustrated in the size of the elements 
required in the electrical equalizer. 

and then transform down again to the microphone impedance. Off- 
hand, it might seem that the transformers required would offset the 
saving in condenser size, but a transformer is normally required any- 
way to bring the very low impedance of the ribbon up to a useful value. 
A third winding has been added to the transformer, so coupled to the 
primary and secondary that a condenser l Aoth the size could be used 
and without any increase in the transformer size. Fig. 14 illustrates 
the reduction in size of the equalizing network made possible by the 
three winding transformer. Thus it is entirely feasible to include 
all of the circuit elements inside the housing of the moving coil ele- 
ment, resulting in a compact instrument. 

We are now in a position to appraise the overall performance of the 
new 6 39- A cardioid directional microphone. The output level is 

268 R. N. MARSHALL AND W. R. HARRY [j. s. M. p. E. 

quite high, 84 db below 1 volt/dyne/cm 2 or 64 db below 1 volt/ 10 
volt/dyne/cm 2 open-circuit voltage across its terminal impedance of 
approximately 40 ohms. This is a level 4 to 5 db higher than that 
of the Western Electric 630-A or 633-A dynamic types already men- 
tioned and only 2 db lower than that of the Western Electric highly 
efficient 618-A dynamic. The normal incidence response of the car- 
dioid combination is smooth over the frequency range from 35 to 
10,000 cycles (Fig. 12), and there is hardly any perceptible quality 
change for any angle of incidence up to 120 degrees. Provided the 
discrimination is good, the quality of the response at the angles greater 
than 120 degrees is of little importance because of the very low sensi- 
tivity in this region. 

For the switch in the D (dynamic) position, the microphone per- 
formance is similar to that of the Western Electric 630-A with the 
acoustic screen removed, both with respect to quality and output 
level, as shown in Fig. 10. While there is a change in quality with 
angle the performance of the dynamic is essentially non directional as 
far as problems of reverberation and feedback are concerned. Since 
this type is so well known in the field, there is no need to elaborate 
here the pick-up technic. Likewise the performance of the ribbon 
element alone, switch in R position, is similar to that of well known 
studio ribbon microphones as far as its bidirectional characteristics 
and quality are concerned. The output level is the same as that of 
the dynamic alone, 90 db below 1 volt/dyne/cm 2 , a level which 
is recognized as being sufficient for general use. 

In order to demonstrate how closely the new microphone follows 
the cardioid directional characteristic at all frequencies, polar curves 
have been plotted in Fig. 15 on a decibel scale. Percentage scales 
are often used for this purpose but are likely to be misleading since 
50-per cent reduction looks like a lot more than the 6 db it actually 
is to the ear. Since the agreement with the theoretical characteristic 
is so good, the cardioid curve may be used for all practical purposes. 

For ready reference, therefore, the chart of Fig. 16, has been pre- 
pared. The figure has been shaded from light to dark to give a 
visual indication of the variation of sensitivity with angle. Also 
several zones have been designated as an aid to remembering how to 
utilize the performance of the microphone to best advantage. A 
"wide pick-up" zone of 120 degrees represents the region in front of 
the microphone where there is practically no variation in quality or 
sensitivity. The "fading" zone from 60 to 150 degrees on either side 



270 R. N. MARSHALL AND W. R. HARRY [j. s. M. P. E. 

indicates that here the sensitivity changes rapidly with angle and care 
must be exercised to keep within range of the microphone. The dead 
zone of 60 degrees at the back of the microphone is that for which 
sounds are discriminated against by approximately 20 db. In addi- 
tion to these three principal zones, the sector from 60 to 90 degrees on 
either side has been selected as a close talking zone. 

To understand the reason behind this close talking zone requires 
a little explanation. The human voice is radiated as a spherical 
sound-wave, for low frequencies at least. Now it is a property of a 
sperhical wave that, whereas the pressure is independent of the wave- 
length, the pressure gradient is not, and becomes proportionately 
very large for points close to the source compared with the wave- 
length. As a result, a microphone of the ribbon type, which responds 
to the pressure gradient, will, when placed close to a spherical sound 
source such as a person talking, favor the lower frequencies of long 
wavelength. This is why ribbon microphones in general use are 
provided with speech "straps" which equalize for low-frequency 
boom on close talking. The necessity for this can be avoided, how- 
ever, in the new microphone by talking at 90 degrees, for in this posi- 
tion the ribbon element is contributing practically nothing, leaving 
the pick-up to the dynamic element which is not affected by the 
spherical character of the sound field. Although this zone is recom- 
mended for close talking, it should be recognized that since the ribbon 
element contributes only part of the output, the accentuation of its 
low-end response by a spherical wave, is only partly reflected in the 
response of the combination microphone. Consequently, even at 
full front the new microphone does not accentuate lows in closely 
delivered speech nearly as much as the ordinary ribbon microphone. 

Exploring of the possibilities of this cardioid microphone in improv- 
ing the technic of sound motion picture recording has only just begun, 
and the results are being awaited with interest. Trials in other 
fields, however, have been so fruitful that three representative set-ups 
will be described here as typical of the results that can be expected 
from the new microphone. 

A small symphonic orchestra of around 30 members was selected 
for one of the studio experiments. The studio selected was one in 
which considerable pick-up difficulty had been experienced, and the 
microphone and orchestra placements are shown in Fig. 17. Note 
that the microphone was set with its dead zone backed near a wall 
leaving plenty of room for arrangement of the orchestra within the 



272 R. N. MARSHALL AND W. R. HARRY [j. s. M. P. E. 

wide pick-up zone. Because of the directional characteristic, reflec- 
tions from the back wall were prevented from interfering with the 
direct sound without having to resort to an acoustic screen or special 
damping of the wall surfaces. Without particular care being paid to 
its exact position, the new microphone handled the situation per- 
fectly although the studio was quite "live" in character. The 
quality of the reproducing sound was characterized by unusual defi- 
nition and sense of space; the brasses, the woodwinds, and the strings 
particularly standing out in full naturalness. Also the bass was 
very rich and clear without being "boomy." This result was di- 
rectly attributable to the true cardioid directional characteristics of 
the microphone. 

The reason for the unusual quality obtained may be attributed to 
three principal features of the new microphone. First, the direction- 
ality gave a control of the balance between the direct and reverberant 
sound so that the individuality of the instruments was not lost in a 
meaningless jumble of reflections. Second, because of true cardioid 
directional response for bass as well as treble tones, indirect sound or 
reflections that were picked up were transmitted with high fidelity. 
Just as a room finished entirely in red would offend our eyes, so would 
pick-up from a microphone which discriminates against all reflected 
sounds except the bass seem boomy; or as a predominant purple 
motif might offend our taste, so would the result from a directional 
microphone which failed to suppress high-frequency reflections grate 
on our nerves. In other words, the quality of the microphone is good 
in the fading zone as well as the wide pick-up zone, so that the total 
pick-up was evenly balanced. 

Furthermore, since the microphone is made up of a pressure and 
pressure gradient element each responding to different properties of 
the sound, the effect of "dead spots" in a studio are practically elimi- 
nated by the use of this instrument. Every room has characteristic 
tones resulting from standing waves which are set up at certain fre- 
quencies by reflections from the walls. Now in a standing wave 
there are nodes at which the pressure is a maximum and the pressure 
gradient zero and vice versa. As a result, these nodes are "dead 
spots" for one type or the other, but in combination both can not be 
"dead" at once. This important feature, no doubt, contributes to the 
"clear bass" observed in the pick-up. 

For another trial, the pick-up of a large symphony orchestra in an 
open air ampitheater was selected. The general layout of orchestra, 


audience, and microphone placement is shown in Fig. 18. Because 
of the large stage and necessity of separating the audience, three 
639- A cardioids were used, two on each side of the front of the stage, 
and a third directly in front of the conductor. This third micro- 
phone covered especially the string instruments and "heard" very 
closely, the same thing as the conductor himself, but the dead zone 
discriminated against accidental noise caused by the conductor. 
All instruments of the orchestra were within the wide pick-up zone of 
one microphone or another without tilting. Furthermore the effect 
of the overhead shell, designed for the benefit of the audience and not 
pick-up, was minimized by the cardioid directional characteristic. 
With this set up it was possible to secure a balance and fidelity of 
pick-up which had not been possible to achieve previously with other 
types of microphones. The enthusiasm expressed by the conductor 
was an indication of the appreciation that fine musicians have for a 
device that enables them to put across their art to the vast unseen 
radio public, without losing the fine details they have worked so hard 
to put into their playing. 

Perhaps the most interesting of the trials, in that it illustrates the 
versatility of the microphone, was in a sound reinforcement system 
for an opera staged in an open-air theater. The stage layout and 
microphone placements are shown in Fig. 19. Four 639- A 's were 
employed, two in the footlights pointing 30 degrees up and two hang- 
ing overhead half way back and pointing degrees 30 down. With 
this arrangement it was found that with fixed levels on the mixer, 
very little variation in quality or level could be observed when a singer 
walked across stage or from front to back. Thus the director was 
informed that he need not instruct the singers to play to the micro- 
phone. To the best of our knowledge this was the first time it has 
been found possible to cover a stage so completely ; and it is suggested 
that this method might be considered for application to certain scoring 
problems. The true cardioid directional characteristic obviously 
made this result possible, the performer moving into the pick-up 
zone of one microphone after another as he walked about the stage. 
Furthermore, the orchestra was in the dead zone of the microphone 
so that there was normally plenty of leeway to balance the singing 
with the orchestra pick-up, and feedback conditions were improved by 
approximately 5 db over that obtainable with other types. Through- 
out the performance the engineer, operating the mixing controls, 
was able to operate the four stage microphones together as if only 

274 R. N. MARSHALL AND W. R. HARRY [j. s. M. P. E. 

one; and thus could concentrate on balancing the singing with the 

The results from the audience viewpoint were startling, some people 
saying that they did not believe the public address system was operat- 
ing when actually reinforcement was practically all they did hear. 
The freedom permitted the performers aided this illustion greatly, 
besides allowing them to act in then: accustomed manner. Again the 
enthusiasm of artists and engineers alike, were a tribute to the unusual 
possibilities in this new microphone. 

These actual trials here described, as well as many others which 
space prevents relating, all testify that pick-up control at the micro- 
phone, a long cherished dream has at last been actually accomplished. 
The key to this control is the directionality for all ranges of the musi- 
cal scale from the lowest bass to the highest overtone. The range of 
this control is from non-directional to bi-directional to cardioid-direc- 
tional performance, but of the three, it is expected that the cardioid 
characteristic may prove to be the most useful form of directivity. 

The secret behind the success of the Western Electric 639-A in 
achieving this true cardioid directional performance lies in the choice 
of the type of pressure and pressure gradient units and the method 
of electrically equalizing and combining the outputs. These also are 
responsible for the high output level of the combination, the ability 
to choose the dynamic, or ribbon units individually, the convenient 
size, and the sturdiness. In addition to its ability to handle any 
situation, to provide control at the microphone, the new 639-A car- 
dioid simplifies the technic of pick-up because of its indifference to 
"dead spots" in the room while at the same time its "dead zone" of 
sensitivity minimizes unwanted slap-back reflections. 


1 MAXFIELD, J. P.: "Some Physical Factors Affecting the Illusion in Sound 
Motion Pictures," /. Acoust. Soc., ID. (July, 1931), pp. 69-80. 

2 MAXFIELD, J. P.: "Acoustic Control of Recording for Talking Motion 
Pictures," J. Soc. Mot. Pict. Eng., XIV (Jan., 1930), pp. 85-95. 

3 Olson, H. F., and Massa, F.: "Applied Acoustics," P. Blakiston's Son fir 
Co. Inc., pp. 136-137. 

4 "A Non-Directional Microphone," Bell Syst. Tech. J. (July, 1936). 

5 STEINBERG, J. C.: "Effects of Phase Distortion on Telephone Quality," 
Bell Syst. Tech J., IX (July, 1930), pp. 550-566. 


MR. KELLOGG: I would like to ask three questions: You have stiffened up 
the ribbon. It is my impression that although this gives you lower resistance it 


does not make up for the reduced amplitude. How much sensitivity did you 
have to sacrifice in stiffening the ribbon? What are the separation and relative 
positions of the pressure and velocity units? Will not the curve be somewhat 
altered if the source of the sound is not in the equatorial plane of the ribbon, or is 
the wavelength below 3000 cycles long enough to make the effect negligible? You 
spoke of damping the ribbon by shunt inductance: I can imagine how shunt in- 
ductance could affect the velocity and phase of the voltage but as a damping 
agent it seems to me that any inductance would be prejudicial. Am I correct 
about that? 

MR. MARSHALL: Regarding the first question, how much loss is obtained by 
stiffening the ribbon, the loss is of the order of 1 db. The curve of sensitivity 
vs. ribbon thickness has a sufficiently broad peak that considerable departure 
may be made from the optimum without taking a serious loss. 

The response of the ribbon unit, since it is essentially a mass-controlled unit, 
is proportional to the weight of the ribbon plus the mass loading of the air. There- 
fore increasing the ribbon alone two and one-half times does not increase the 
effective mass that much, and at the same time the electrical resistance is also 
reduced. There are also other factors that have to be taken into account, such as 
the leakage path around the edges of the ribbon and the width of the baffle path 
and so forth. I would say that the formula we used for calculation shows a theo- 
retical loss of about 0.9 db and we have checked that pretty closely. 

As to your second question "How did the directivity vary in the vertical 
plane" it is well known that the separation of the microphone units will cause a 
phase shift between them at high frequencies. In this particular microphone 
the separation is approximately l x /2 inches, I believe. As you yourself intimated, 
the fact that we are filtering out the ribbon unit at 3000 cycles enables us to re- 
tain directivity that is more uniform than one might expect. The directivity in 
the horizontal plane does not vary seriously. With respect to the vertical plane 
it does vary a little. The phase shift we do get will either be helpful at some point 
and give better directivity, or be harmful and give less. The average is about 
the same in this particular unit. 

As to how the ribbon becomes damped by the use of shunt inductance, the 
shunt inductance does two things. Not only does it introduce electrical loss but 
at the same time allows current to flow back through the ribbon. It is a sort of 
feed-back principle in which the current that goes through the inductance regu- 
lates the motion of the ribbon. There is also some electrical resistance which is 
reflected back, through the electromagnetic coupling, into the mechanical side. 

MR. KELLOGG : Is not the presence of the inductance prejudicial to the damp- 
ing of the amplitude of the ribbon? 

MR. MARSHALL: No; as can be shown theoretically and practically. One 
may call it prejudicial, but it does perform in a uniform manner. I have not ob- 
served anything undesirable from shunting. 

I tried to outline a moment ago that we have a way of regulating some of the 
output. The equations show that the magnetic coupling factor is sufficient to 
bring down the response to a point where it is satisfactory and can be controlled. 
As a matter of fact, the size of the inductance has to be controlled to take care of 
variations in flux density, and we get a proper coupling factor which reflects back 
the right amount of this so-called feedback current and resistance. It is quite 

276 R. N. MARSHALL AND W. R. HARRY [j. s. M. P. E. 


complicated. There is actually a resistance in this inductance circuit as well as 
feedback current. There is also electrical loss and all three factors are concerned 
in the problem. 

MR. STEVENS: Is it impossible to obtain at high frequencies better directivity 
than shown by your curves? 

MR. MARSHALL: No, but this was the best we were able to do with this par- 
ticular microphone at this time. 

MR. STEVENS: In your opinion is it possible to provide more? 

MR. MARSHALL: It is possible, but we run into factors at high frequencies 
that make it very difficult, such as diffraction effects and phase-shift caused by 
separation of the units. 

MR. WILLIAMS: I assume that the curves you show were taken in a plane 

MR. MARSHALL: There is no such thing, practically. Field calibrations have 
been standardized in terms of plane waves and, of course, we approximate a plane 
wave the best we can. 

MR. WILLIAMS: I believe I am right in saying that there is considerable dif- 
ference between a plane wave and a spherical wave near the source. You have 
corrected for the two and added cancellation from the back in the plane wave. 

MR. MARSHALL: That is correct. 

MR. WILLIAMS: You have demonstrated what part the spherical wave plays. 
Do you get equally good cancellation in the plane wave? 

MR. MARSHALL: We get better. 

MR. KIMBALL: What is the variation in output level in the three switch 

MR. MARSHALL: A ratio of two to one. Naturally, with one unit we get 
half as much voltage. 

MR. STANCIL: When the switch is on the cardioid position, a low-pass filter 
cuts in above 3000 cycles attenuating the ribbon element. Now, if this switch 
is turned to the ribbon position, would the filter still be in the circuit, attenuating 
much of the high-frequency response? 

MR. MARSHALL: No. The switch when turned to ribbon position also re- 
moves the filter or equalizer circuit. 

MR. HILLIARD: What is the characteristic of the non-directional unit com- 
pared with that of the 630 mike. 

MR. MARSHALL: It is similar to the 630 without the screen. The same dy- 
namic unit that is used in the 630, is used in the 639-A without alteration or change. 

MR. THAYER: Does the difference in the front of the moving-coil housing of 
the 630 and 639 have any effect on the response? 

MR. MARSHALL: Yes. If you will recall, the curve I showed of the dynamic 
action was the curve of the 630 microphone without screen. The acoustic screen 
of the 630 keeps out the residual directional effect at high frequencies and if you 
remove it, the microphone becomes semi-directional. 

MR. THAYER: The diaphragm housing of the 639 mike is different from that 
of the 630? 

MR. MARSHALL: There is a difference but the difference is not such as to affect 
the directional pattern seriously. The major portion of the directional effect in 


this microphone is due to the physical size of the diaphragm, which is about 
l l / 4 inches in diameter. 

DR. DAILY: On account of the physical separation between the velocity and 
pressure elements of this type of microphone, there will be peaks and valleys in the 
response characteristic, their amplitude depending to some extent on the angle of 
incidence of the sound-wave. Is there an appreciable increase in the differential 
amplitude of these peaks and valleys as the angle of incidence changes from nor- 
mal to 35 or 40 degrees? 

MR. MARSHALL: There will not be much effect until you get around to the 
back. Up to 90 degrees it might be about 2 db. At the back the variation 
may be as much as ==4 db. 



Summary. Factors influencing the choice of a microphone for sound recording 
are considered. The characteristics of a new miniature condenser transmitter and 
amplifier, as well as a number of other types of microphones now in use, are included. 

One of the most interesting problems with which the sound engi- 
neer has to contend in the communication, broadcasting, and sound 
recording fields, is the provision of a suitable instrument to translate 
the acoustic variations in speech and music to corresponding electrical 
variations. The microphone performs this function, and the pur- 
pose of this paper is to consider its application to sound recording for 
motion pictures. 

Requirements for microphone performance hi sound recording are 
somewhat more critical than in other fields. This is due to the variety 
of pick-up conditions encountered, and to the necessity of listening to 
the recorded material many times during the course of production. 
In dialog recording, the microphone position is changed almost con- 
tinuously to allow wide scope of action on the set. Set construction, 
lighting, and camera angles necessarily limit a choice of microphone 
position and often make it difficult to secure an optimum position. 
These restrictions become less severe if the microphone is sufficiently 
small so that it may be easily kept out of the field of view of the 
camera. It should be light in weight so that it may be used inter- 
changeably on a microphone boom or upon a fishpole when space 
requirements restrict the use of a boom. A light-weight microphone 
is more easily handled for action shots where rapid movements of 
the microphone are required to cover action properly. Its use for 
this latter type of pick-up predicates freedom from wind noise and 
mechanical shock. Dialog recording on location or for newsreel 

* Presented at the 1 939 Spring Meeting at Hollywood, Calif. ; received April 
13, 1939. 

** Electrical Research Products, Inc., Hollywood, Calif. 



service requires the instrument to be sufficiently rugged to withstand 
mechanical shock, and to be relatively unaffected by wind, altitude 
changes, temperature variations, or moisture. 

Program material in sound recording is not transitory in char- 
acter. One scene may be repeated a dozen times for the benefit of 
action or photography, and the subsequent reproduced recordings 
must of a necessity be judged for acceptable sound quality in the re- 
view room. By having an opportunity to become familiar with the 
quality of the actor's voice, both by direct listening on the set and 
from the reproduced recordings, a fairly critical estimate can be made 
of this quality throughout the picture. If different microphones, or 































*< / 



FIG. 1. Equalization characteristic for sound recording. 

different types, are used from day to day, any variation in quality 
due to microphones is easily perceived. This, of course, places a 
premium on similarity of response characteristics of various micro- 
phones of the same type, and upon their ability to maintain such a 
characteristic in service. 

Another limitation is the present-day use of monaural sound sys- 
tems. 1 Listening with both ears, binaurally, the sound-sources can 
be localized, and the interfering effects of background noises and 
reverberation minimized. The microphone inherently can do neither. 
To compensate for these deficiencies, present technic requires that 
the microphone be placed close to the sound-source to minimize these 



[J. S. M. P. E. 

disturbances, or the use of an instrument having marked directional 

With some of these factors in mind, consideration may be given to 
the requirements for an ideal microphone. In the first place, the 
output should contain all the frequency components present in the 
original with comparable amplitudes, and within certain limits bear 
the same phase relationship to each other. It is essential that the 
microphone's electrical output be sufficiently high so that the noise- 




& -95 












* % \ 






r \r./s 















. f 

* /- 








\ ^ 


































FIG. 2. 6 18- A microphone characteristics. 

level of the system to which it is connected does not become a limit- 
ing factor. Its directional characteristics are not as easily defined 
since various degrees of directivity may be required. Size and weight 
play an important part in its ease of handling on the stage, and sus- 
ceptibility to such interferences as wind, movement, and power ex- 
posure, must be considered. 

Present-day designs approach these fundamental requirements, 
and offer instruments which may be successfully adapted to sound 
recording. It might appear that with such devices it would be neces- 


sary only to connect them to a recording channel and proceed. We 
find, however, that due to factors such as set acoustics, differences 
in the actor's speech volumes as originally heard and subsequently 
reproduced, and certain considerations of the recording and re- 
producing system characteristics, some modification of the micro- 
phone response characteristic is required. This is usually accom- 
plished by electrical equalization in the recording system, or some- 
times by the addition of devices to the microphone for acoustically 
altering its response. While such equalization necessarily varies 
with the type of microphone used, and class of recorded material, 
it usually falls within the limits indicated in Fig. 1 . The two sets of 
limits correspond to those employed for dialog and musical recording, 
and differ principally in the amount of attenuation used at low fre- 
quencies. The attenuation of low frequencies in dialog recording is 
occasioned by set acoustics and volume differences in the original 
and reproduced material. By set acoustics is meant reverberation, 
the possibility of first-order reflections from walls or objects on the set, 
and vibration of insecurely braced walls or set materials. Volume 
differences in original and reproduced dialog are occasioned by the 
higher volumes required in the theater for satisfactory reproduction, 
and due to the non-linear characteristic of the ear, represent an in- 
crease in low-frequency response which is compensated for by equali- 
zation in recording. The indicated increase in response in the region 
of 2500 cycles is sometimes employed to increase "presence." "Pres- 
ence" refers to the completeness of illusion that the voice from the 
theater screen is emanating from the pictured actor. This results 
in a feeling that the sound is being reproduced intimately at the 
screen, and not at some location remote to it. 

Microphones now in use generally belong to one of two groups: 
the electrodynamic, including the various pressure-operated moving- 
coil types, and pressure gradient ribbons; or the electrostatic de- 
vices, represented by the condenser transmitter. The salient proper- 
ties of those in common use, as well as some of the more recent types 
are given in Table I. 

Extensive descriptive material has been published covering the 618 
and 630 type microphones. 2 ' 3 The 639 type has been described fully 
by Marshall and Harry. 1 The 640 -A transmitter is an improved 
miniature condenser. 4 Its use, with the RA-1095 transmitter am- 
plifier, has not been previously described and will now be considered. 
Condenser microphones have been used in one form or another since 



LJ. S. M. P. E. 












dance Weight 

Dynamic mov- Fig. 2 Semi-direc- 83 db 30 ohms 
ing coil tional at high 




Dynamic mov- 
ing coil with 

Fig. 3 Non-directional -89 db 20 ohms 1 Ib 

630-A Dynamic mov- Fig. 4 Semi-direc- 89 db 20 ohms 1 Ib 

ing coil with tional at high 

flat baffle frequencies 

639- A Combined Fig. 5 U n i - d i r e c 84 db 35 ohms 3 V4 Ibs 

"C" moving coil tional Car- 

and ribbon dioid 



Bi-directional 90 db 30 ohms 3 l / 4 Ibs 

640-A Condenser Fig. 6 Semi-direc -- 61 db 50 ohms 

tional at high 

RA-1095 frequencies 


*0 decibel = 1 volt per bar (open circuit). 




E 0= SC ^E-JCE 







' ' i i 
B B B 5 1 









. V 



1 1^ 


Z ^60 

~- " 

*" K. 

\\ I8 



30 50 

500 KX 


5000 tOfXX) 20,000 

FIG. 3. 630-A microphone with screen. 




Z -90 

r 95 
















s ^ 















toe 500 


FIG. 4. 630- A microphone with a 3 'A baffle. 


FIG. 5. Laboratory model of 6 39- A cardioid directional microphone. 



IJ. S. M. P. E. 

the inception of sound recording and, prior to that, in communica- 
tion systems. With the development of the smaller and lighter 
moving-coil microphones, their use has gradually diminished. Now, 
due to recent developments of the Bell Telephone Laboratories in 
providing a miniature transmitter, vacuum tube, and output trans- 
former, a design becomes possible that can compete with the moving- 
coil type with respect to size and weight. The condenser transmittei 
has an inherent advantage over the moving-coil microphone in that 
its mechanical structure is relatively simple. As a result, it possesses 
a smoother response characteristic than has been practicable in the 
moving-coil type. For the same reason it has been possible to main- 
tain a much greater degree of uniformity among various individual 
condenser transmitters, than between various moving-coil micro- 



| -65 



FIG. 6. RA-1095 amplifier with 640- A condenser transmitter. 

phones of the same type. The high quality of performance of whic 
this microphone is capable, together with its convenient physic; 
form, promises to re-establish as a recording tool. 

The appearance of the amplifier and associated condenser tran:' 
mitter is shown in Fig. 7. The microphone amplifier is about 7 inch< 
long, with a maximum diameter of 2 l / z inches, and weighs I 1 .! 
pounds. The tapered shell housing is easily removed from the chass 
by loosening two screws at the base, and sliding it from the chassi 
The chassis consists of a metal frame carrying all equipment. Tl 
vacuum tube and transmitter coupling mesh are mounted on mat 
rial possessing high insulation resistance, and a number of precaution 

Sept., 1939) 



have been taken both to establish and maintain it. The vacuum 
tube is of the heater type designed for long service, and is soldered 
into its socket, which reduces contact troubles and helps to maintain a 
high degree of insulation. These soldered connections represent no 
hardship in use, as it is usual studio practice to replace defective 
microphones on the stage with others, and to repair them elsewhere. 
A terminal strip on the base of the chassis permits changing output 
impedance, and ground connections to operate under varying studio 

The combined acoustical and electrical characteristics of the 640-A 
transmitter and RA-1095 amplifier are shown in Fig. 6. This char- 
acteristic is a combination of the acoustic response of the transmitter, 
any effect that the amplifier housing may have upon the trans- 
mitter's response due to its modification of the sound field, and the 

FIG. 7. 640-A transmitter and associated amplifier. 

amplifier's electrical characteristic. Since the amplifier frequency 
response is essentially uniform over the range of recorded frequen- 
cies, it may be considered as a means of coupling the transmitter to its 
associated recording circuits. Previous laboratory data indicate 
that the characteristic illustrated is essentially that of the transmitter ; 
hence we may conclude that the amplifier housing does not appreci- 
ably effect the transmitter response. The rise at high frequencies 
in the normal incidence response is caused by the diffraction of sound 
by the microphone. Diffraction effects, however, are fairly small for 
angles of incidence greater than 60 degrees, and it is the usual practice 
to use the microphone in such a manner that most of the desired 
sound is incident at these angles. Referring again to Fig. 6, the 
response characteristics for angles of incidence of 120, 150, and 180 
degrees, are essentially those shown for 90-degree incidence. The 
output of the amplifier is approximately 61 db/one volt per bar, 



LF. S. M. P. E. 

open circuit. It is about 24 db greater than the output of the 630-A 
microphone. The noise output compares favorably with that of 
similar amplifiers employing large tubes especially selected for low- 
level circuit application. Numerous recording and listening tests 
have indicated that this instrument provides improved overall qual- 
ity, greater naturalness, and intelligibility, compared with older types 
of microphones. 

For stage pick-up microphones are usually supported by means of 
fishpoles or microphone booms to permit rapid movement of the 
microphone about the set to cover the action properly. As a result 
the microphone is subject to considerable mechanical vibration, and 
some form of insulating mounting is required to prevent these vibra- 
tions from being transmitted to the microphone and thus recorded. 

FIG. 8. Plate type of mounting, adapted to 630 micro- 

Two types in common use are the double-ring and the double-plate 
forms. In the first a double-ring mounting is used, the inner ring 
being fastened to the outer by means of elastic bands. The micro- 
phone is then clamped in the inner ring, the outer being fastened to 
the boom head or fishpole. 5 This form of mounting is applicable to 
microphones having cylindrical housings, such as the 618 or RA-1095 
condenser transmitter amplifier, or with an adapter to the 630 type 
microphone. The plate type of mounting adapted to a 630 micro- 
phone is shown in Fig. 8. In this type two plates are isolated from 
one another by means of Lord rubber mountings. The microphone is 
carried by one plate, the other being attached to the boom. 

When microphones are used in the presence of wind, slight changes 
in wind velocity cause corresponding changes in pressure at the face 

Sept., 1939] 



of the microphone, resulting in a low-frequency disturbance in the 
microphone output. The amplitude of the disturbance is often suf- 
ficiently high to interfere seriously with the recording. A wind- 
screen or bag affords a method of minimizing this pressure change, 
and is quite effective in increasing the microphone's usefulness under 
such adverse pick-up conditions. The wind-bag in its simplest form 
consists of a wire screen cage surrounding the microphone. It is 
usually covered with two layers of porous silk separated by a small 
air space. It effectively forms a high-pass acoustic filter and attenu- 
ates the very low-frequency wind- 
pressure variations. Over the re- 
cording range, it has little effect 
upon the microphone response 
characteristic. A typical wind-bag 
for use with the 630-A microphone 
is illustrated in Fig. 9. 

Maintenance of microphone 
equipment in a studio includes a 
periodic check of the response 
characteristic and volume output. 
This is usually accomplished by 
comparing a standard calibrated 
microphone of the same type, with 
the one under test. The measure- 
ment is sometimes made as a lis- 
tening test, with both instruments 
being alternately connected to a 
monitoring system, or, more fre- 
quently, some sort of acoustic 
measurement employing a sweep 
oscillator, and level recorder, from 

whose charts comparison is made between the calibrated microphone 
and the one under test. Another common problem where micro- 
phones having magnetic structures are used, is that of removing 
iron particles picked up by the unit. By the use of a brush and 
some form of sticky tape, it is usually possible to remove them. 

It has been the object of this paper to indicate a number of factors 
that materially influence the choice and operating conditions for a 
microphone for sound recording. The ckaracteristics of a number of 
commercial types in use have been shown, and a brief description 

FIG. 9. 

Wind-bag used with 630-A 



has been given of a new miniature condenser transmitter and amplifier 
giving improved performance. 


1 MARSHALL, R. N., AND HARRY, W. R. : "A Cardioid Directional Micro- 
phone," /. Soc. Mot. Pict. Eng., XXXTTT (Sept., 1939), p. 254. 

2 JONES, W. C., AND GILES, L. W.: "A Moving Coil Microphone for High- 
Quality Reproduction," /. Soc. Mot. Pict. Eng., XVII (Dec., 1931), p. 977. 

3 MARSHALL, R. N., AND ROMANOW, F. F. : "A Non-Directional Microphone," 
Bell Syst. Tech. J. (July, 1936), p. 405. 

4 HARRISON, H. C., AND FLANDERS, P. B., "An Efficient Miniature Condenser 
Microphone System," Bell Syst. Tech. J. (July, 1932), p. 451. 

5 STROCK, R. O., "Some Practical Accessories for Motion Picture Recording," 
J. Soc. Mot. Pict., Eng., XXXTT (Feb., 1939), p. 188. 


luminary. After an explanation of the term Class A-B and a brief specification 
of such a recording system, the general requirements for the operation of any Class A-B 
system are given and illustrated. 

Differences between the operation of push-pull photocells and push-pull vacuum 
tubes are pointed out and explained, and a discussion of the relative advantages of 
Class A, Class A-B, and Class B push-pull tracks is given. 

A push-pull sound-track consists of two simultaneously recorded 
tracks that are mutually related so that the original sound can be 
reproduced by means of a push-pull reproducing system. 

A push-pull reproducing system consists of essentially two photo- 
cells for detecting the signals recorded on the two component parts of 
a push-pull track. The two photocells are both connected to a 
sound producing system and are mutually 180 degrees out of phase. 

For a push-pull Class-A sound-track each of the two component 
| tracks are complete records which are recorded 180 degrees out of 
phase. When the detecting system is in perfect balance, all even- 
harmonic distortions produced by processing of the two component 
tracks are automatically eliminated. When the detecting system is 
j not in perfect balance, even-harmonic distortions are still partially 
compensated and since both component tracks are complete records 
no additional distortion is introduced due to the unbalance; simply 
an attenuation of the signal strength occurs. (This attenuation may 
i selective depending on the nature of the unbalance.) As with 
standard recording, the ground-noise reduction is accomplished by 
means of automatically operated shutters. Ideally any modulation 
produced by movements of the shutter vanes should be automatically 
eliminated in the push-pull reproducer. Actually at present it is 
possible only about to double the shutter speed over that of a stand- 

* Presented at the 1939 Spring Meeting at Hollywood, Calif.; received April 
' 9 j-v/oy. 

* Massachusetts Institute of Technology, Cambridge, Mass. 
RCA Manufacturing Co., Hollywood, Calif. 

290 C. H. CARTWRIGHT AND W. S. THOMPSON [j. s. M. p. E. 

ard recording before noticeable distortion occurs. Due to this gain 
in shutter speed the clearance lines are reduced to half the width of 
that used for standard recording and hence the ratio of signal to 
ground noise is increased by about six decibels. 

For a push-pull Class B sound-track, only deflections of the 
recording galvanometer in one direction from its mil position are 
recorded on one of the component sound-tracks; deflections in the 
other direction from the galvanometer's mil position are recorded on 
the other component sound track. Since the two component records 
must be combined perfectly to make a complete record, any unbal- 
ance in the reproducing system creates distortion. The main feature 
of the push-pull Class B sound-track is its inherent noise reduction. 
No shutters are required for noise-reduction and the ratio of signal 
to ground-noise is higher than for any other system of recording. 
The disadvantages of Class B recording are: (1) the necessity of a 
well balanced reproducing system, (2) the extreme sensitivity of the 
azimuth adjustment for the recording and reproducing slits, and (3) 
the critical requirements for processing the film and the dependence 
of the correct azimuth setting of the recording aperture on the film 

A push-pull Class A-B sound-track consists of a pure Class A 
record for low modulations up to a predetermined level and a com- 
bination of Class A and Class B records for higher modulations. The 
features of this type of recording are : 

(1) Ground-noise reduction is inherent in the record; 

(2) Azimuth adjustments in recording and reproducing are less critical than 
for Class B recordings; 

(5) Film processing is less critical than for Class B recordings; 
(4} Balance in the reproducing system is less important than for Class B re- ; 

In the following description and analysis of Class A-B recording, 
only variable-area sound-tracks will be treated explicitly; however, 
there is an intimate correspondence to Class A-B variable-density 
sound recording which should be apparent. 

General Requirements for a Class A-B Recording Aperture. Due 
to the general familiarity with vacuum tube amplification, it may be 
well to point out the differences which exist between Class A-B 
vacuum tube amplification and Class A -B sound detection by the use 
of photocells. Fig. 1 illustrates the relationship between input and 
output voltages to be fulfilled by any push-pull Class A-B system. 



For the pure Class A amplification both vacuum tubes or photocells 
are equally effective in producing an output voltage. Beyond the 
pure Class A portion only one device is operative and, hence, must 
yield twice the output per input ratio as it did for pure Class A. 
Efficiency considerations make it desirable to match impedances and 
in the case of the vacuum tubes this impedance matching can be at- 
tained and it automatically yields the ideal relation between input 
and output as shown in Fig. 1, except for a small distortion due to 
the toe characteristic of the vacuum tubes. That is, when one 


FIG. 1. Input versus output relation to be fulfilled by any 
push-pull Class A-B detecting and amplifying system. 

vacuum tube becomes inoperative, as is the case beyond the pure 
Class A amplification, the efficiency of the other is automatically 

In the case of photocells, the desired impedance matching is not 
attained and the efficiencies of the two photocells are practically inde- 
pendent. This means that beyond the pure Class A region the 
effective output per input ratio of each photocell acting alone must 
be doubled by some means. The means for doubling the effective 
output per input ratio which we have employed has been the use of a 
special recording aperture such that the change in the track width 



per change in the input to the recording galvanometer is half as great 
for the pure Class A record as for the region beyond the pure Class A 
record. Thereby the desired relation in Fig. 1 is fulfilled between the 
input to the galvanometer and the output of the push-pull reproduc- 
ing system. Fig. 2 shows the relationship to be fulfilled between the 
deflection of the recording light-beam (which is proportional to the 
input to the galvanometer) and the width of the two component 
sound-tracks produced. Fig. 3 illustrates a Class A-B recording 
system. The requirements described in Fig. 2 can be fulfilled by 

FIG. 2. Relation to be fulfilled between width of sound tracks and the deflec- 
tions of the recording light-beam by a Class A-B recording aperture. 

several different apertures. The Class A-B shown in Fig. 4 could be ; 
produced by the aperture shown in Fig. 5. 

Detailed Description of Class A-B Recording Apertures. Fig. 5 : 
shows a Class A-B recording aperture which at full modulation gives 
two 35-mil component sound-tracks on the film with a minimum 
separation of 6 mils between the two component tracks. The sensi- 
tivity of this aperture is such that for full modulation the galva- 
nometer deflection is the same as now being used in all standard equip- 
ment. The choice of dimension a or b depends on the proportion of 
pure Class A recording desired which, in turn, is proportional to the 
amount of ground-noise reduction inherent in the track produced. 



Tails each 2 mils wide (about y 4 mil on film) are provided to lessen 
distortion due to processing. In Table I are data which show differ- 
ent values of a, the corresponding value of b, the inherent amount of 
ground-noise reduction expressed in decibels over a standard system 
used without any ground-noise reduction, the proportion of full 
modulation that is a pure Class A expressed in decibels, and the 
width of the zero lines produced on the sound-track film. 

FIG. 3. Class A-B recording optical system. 


Class A-B Recording Apertures 

Noise Reduction End of Class A 

(db) (db below 100%) 

12.5 -28 

11.5 -25 

10.5 -22 

9.5 -21 

9.0 -19 

8.0 -16.8 

7.0 -14.8 

6.3 -13.2 

5.5 -12.0 

Width of Track with 
No Modulation (Mils) 



Azimuth, Modulation, and Distortion Tests of Class A-B Records. 
An aperture as shown in Fig. 5 with the dimensions a equal to 15 
mils was constructed and used for the tests to be described. For 
this aperture all modulations of less than about 15 db from full modu- 
lation produced pure Class A records. 



Using a 400-cycle pure sine- wave signal a series of records was 
made for various azimuth settings and for various amplitudes. The 
harmonic distortion for these settings was measured by a distortion- 
factor meter connected to the output of a reproducing system. 

The results of these tests show that the azimuth settings for Class 
A-B recording are much less critical than for Class B and that they 
are most critical for those levels corresponding to the level at which 
the Class B portion of the track is just starting. The limits of azi- 
muth setting were such that azimuth could be set by visual inspection 
to yield a satisfactory recording from a distortion standpoint. 

With the correct azimuth setting, harmonic distortion was mea- 
sured for various signal levels. These tests showed very low harmonic 
content for both the Class A and Class A-B positions of the track and 
an increase of only 0.15 per cent in harmonics at the transition point. 

FIG. 4. Class A-B recording sound-track. 

Conclusion. The inherent advantages of a push-pull reproducing 
system can be gained by means of a Class A-B sound-track; it has 
the advantage over pure Class A recording in that the ground-noise 
reduction is inherent in the track and does not require the use of 
automatically operated shutters and it has the advantage over pure 
Class B recording in that low-level modulations are not distorted as 
much by film processing or a faulty azimuth setting of the recording 
aperture, the recording slit, or the reproducing slit. 

The amount of ground-noise reduction inherent in a Class A-B 
record is inversely proportional to the ratio of pure Class A record 
to the full track record. 

The dimensions of Class A-B aperture giving as much inherent 
ground-noise reduction as is now obtained by the use of shutters for 
standard bilateral recording or Class A push-pull recording are such 
that the construction of the aperture is practical and the resulting 
proportion of the track that yields a pure Class A record is sufficient 
to remove the excessive distortion of low-level modulations sometimes 
found in a Class B record due to film processing and azimuth settings. 



An analytical treatment of the distortions introduced by faulty 
azimuth settings for sine waves reveals the distortions of high-level 
modulations to be practically the same as for a pure Class B record. 
However, the low-level modulations of the Class A-B record that 
are pure Class A records are not distorted by a faulty azimuth setting, 
while the percentage distortion of a Class B record increases as the 
intensity of modulation decreases. An unbalance in a reproducing 
system distorts all modulations of a Class B record equally, but 
in a Class A-B record there is no distortion in the Class A portion of 
the record and the distortion for larger modulations is less than for a 
Class B record. 

As far as inherent noise-reduction is concerned the Class B track is 
the quietest known to date and is the ultimate toward which we 

FIG. 5. Class A-B recording aperture for standard 
modulation efficiency. All dimensions in thousandths of 
inch on aperture. 

lould progress. However, under certain adverse conditions there 
are phases of Class B recording which become critical, such as film 
processing, reproducer balance, etc. The Class A-B record is not 
as free from these conditions as the Class A track but is less critical 
than the Class B record and has the advantage of inherent noise- 
reduction without the use of shutters and noise-reduction amplifiers. 
It is possible at the present time to obtain noise-reduction equal to 
that of our standard and Class A tracks and we see the possibility 
of progressively using less and less Class A portion as conditions 
stabilize until straight Class B tracks can be used. 

While we have treated only a variable-area type of sound-track, 
the same considerations are obviously applicable to a variable- 
density type of sound-track. Several methods immediately suggest 
themselves for producing a variable-density push-pull Class A-B 
sound-track such as the use of a special light-valve, a special 
penumbra system, or a pair of special filters for the recording aperture. 


Summary. A fundamentally new principle in desgn of photographic developers 
has been investigated and found to afford many worthwhile characteristics, chief of 
which is the effective self-replenishing property of the developer solutions. Applica- 
tion of the new principle to developer solution makes it possible to develop about eight 
times the quantity of film as would be possible under ordinary conditions. The prin- 
ciple may be applied to any developer. 

As a brief preface to these remarks on the subject of aluminate 
developers, the following paragraph by J. I. Crab tree and C. E. Ives 
will be appropriate. 

The successful operation of a motion picture laboratory depends very largely 
upon its ability to produce prints of uniformly good quality. Such prints can be 
produced only by maintaining constant the various factors which control the ex- 
posure and degree of development of the image, and of these the developing power 
of the developing solution is perhaps the most difficult to control. 

The RCA Manufacturing Co. has spent considerable time and re- 
search in an effort to improve the sound quality of final release prints. 
First attempts were made in cooperation with film manufacturers to 
reduce the image spread of a given film emulsion and thereby in- 
crease the tolerances of a negative-print density combination. 
Research on the part of film manufacturers has in the past few years 
resulted in the availability of sound recording emulsions having 
greater speed without sacrifice of resolution or increase of image 
spread, and of emulsions for other purposes which represent great 
improvement over previous emulsions. On our part a study has 
been made of developing solutions with a view of obtaining the de- 
sired increased processing tolerances. Although the study of de- 
velopers and characteristics of developing solutions was entered upon 
primarily for the purpose of improving the quality of sound-on-film 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received March 
31, 1939. 

** RCA Manufacturing Co., Camden, N. J. 


recording, it became increasingly evident during the course of inves- 
tigations that there existed little-known and little-realized principles 
of chemical behavior which broadened the problem from the restricted 
application to sound recording to a general overall improvement in 
developer performance. The developing power of any ordinary devel- 
oper begins to diminish the moment film is introduced into the de- 
veloper, due to the release of reaction products which exert restraining 
action, along with the fact that a certain amount of the developing 
agent is used up. In order to overcome this difficulty with processing, 
laboratories have various methods for adding a replenisher to the 
developer as film is being developed, in an attempt to maintain con- 
stant density for a given exposure at a given developing time and 
temperature. To understand the problems involved in an endeavor 
to improve developer characteristics, the general theory of develop- 
ment should be considered. 

Ordinarily there are four essential ingredients in a developer : 

(a) The organic reducing agent 

(6) The ionizing agent or the energizer 

(c) The preservative 

(d) The restrainer 

The function of these ingredients is evident. The effect of varying 
their concentration is, in general, as follows: Increasing the concen- 
tration of the reducing agent shortens the development time. In- 
creasing the concentration of the ionizing agent may have a number 
of consequences. First, it increases the energy of the developer, not 
only shortening the developing time but also increasing the tendency 
toward fog. Another effect is the physical effect on the gelatin of the 
film by the ionizing agent, which is usually an alkali, causing the 
gelatin to swell. Altering the concentration of the preservative also 
may have a number of consequences. At high concentration of 
sodium sulfite, which is used as a preservative, there arises an effect 
on the silver halide grain in the emulsion whereby the silver salts tend 
to go into solution. The effect of such solvent action is to decrease 
the contrast and to lower the effective emulsion speed of the film. 
The reason for this is evident inasmuch as during the course of de- 
velopment portions of the silver halide crystals are dissolved away 
before they can be developed to metallic silver. 

The restrainer is generally potassium bromide. Its presence in the 
solution has an effect of inhibiting the fog and lowering the toe den- 
sities. Increasing the concentration up to a certain point gradually 

298 J. R. ALBURGER [j. s. M. P. E. 

increases the restraining action. Wlien this point is reached corre- 
sponding to a point of about 0.08 mol, any further increase of the 
concentration of potassium bromide will have little effect on the 
characteristics of the developer. 

It is true without doubt there are few things that are perfect, and 
few things the improvement of which would not be greatly desired. 
In the case of developers a number of desirable improvements can be 
enumerated as follows : 

(a) Increased stability 

(6) Increased life of developer 

(c) Higher possible contrast 

(d) Improved control of developed contrast and emulsion speed 

(e) High resolution 

(/) Higher emulsion speed 
(g) Lower fog 
(ti) Lower cost 

To increase stability a number of possibilities present themselves. 
First of all, the bromide effect should be noted. During the course 
of development, bromide is released from the silver salts in the emul- 
sion. The presence of the bromide thus released tends to inhibit 
or slow down subsequent development. A practical way to meet 
this difficulty and to improve stability so far as bromide effect is 
concerned, was found in simply starting with sufficient bromide 
already in solution so that the slight increment of bromide added 
during development would not cause an appreciable progressive 
slowing down of the developer. Fortunately it was found possible 
to formulate a developer the energy of which was sufficiently great so 
that the presence of large quantities of potassium bromide would not 
decrease its developing energy below a useful point. It was found 
that a concentration of potassium bromide of between 6 and 10 
grams per liter was optimum. The life of a developer might be im- 
proved by increasing the concentrations of essential ingredients but 
there would still be present instability represented by exhaustion. 
If the concentration of the developing agent were to be doubled, 
we should expect the developer to last about twice as long, but the 
exhaustion characteristic has not been changed by this change in 
concentration, and instability from this source would still be present. 


Now we enter upon what is believed to be a fundamentally new 
function as applied to developing solutions. The chemistry of this 


function is not new but its application to photography seems to have 
distinct novelty. 

Certain metals when treated with sodium hydroxide behave in the 
following manner: First, the insoluble hydroxide of the metal is 
precipitated and further addition of sodium hydroxide will re-dissolve 
this precipitate. The metals that behave in this manner are few. 
They are aluminum, lead, tin, zinc, and chromium, and the rarer 
metals, indium, gallium, and germanium. 

With aluminum as an example, the reaction would be as follows : 

A1 2 (S0 4 ) 3 + 6 NaOH -* 3Na 2 SO 4 + 2A1(OH) 3 J, 

[+6NaOH 2f 2Na 3 AlO 3 + 6H 2 O + 3Na 2 SO 4 ] (2) 

The theory set forth in chemistry textbooks for the re-solution of 
these metallic hydroxides is that a polypeptized sol is produced. 
The terms peptize and peptization are generally used in reference to 
organic chemistry where the chemical function thus designated is the 
building up of large molecules from a number of smaller molecular 
groups. We are most familiar with peptization in its occurrence in 
the human body where the catalyst "pepsin" present in the digestive 
system serves to synthesize large complex molecular structures from 
the component chemicals of the foods we eat. In applying the idea 
of peptization to the aluminate developer a somewhat related action 
takes place. A large number of sodium aluminate molecules become 
grouped together in such a way that a large complex molecule is 
formed. This complex molecule is found to be soluble. The mecha- 
nism of this peptization is believed to be as follows: Aluminum hy- 
droxide exists in equilibrium with acid aluminate. The aluminum 
hydroxide is an insoluble gelatinous material. The theory is that 
in the presence of an excess of hydroxyl ions a number of acid alumi- 
nate molecules become grouped together and peptized in combination 
with a molecule of sodium aluminate. To illustrate, the formula for 
aluminum hydroxide is 

| A1(OH) 3 *=; H 3 A10 3 (2} 

This equation represents an equilibrium wherein an alkali is balanced 
against an acid. The presence of additional alkali will shift the 
equilibrium point to the right. 

It should be remembered that there is at all times a reaction equi- 
librium existing in the developing solution. In such a reaction 

300 J. R. ALBURGER [j. s. M. P. E. 

equilibrium, any change in the concentration in any of the reaction 
products will shift the balance point of the equilibrium. Thus, if 
the metallic hydroxide is dissolved in the sodium hydroxide in such a 
manner that there is just sufficient hydroxide to hold it in solution, 
and no more, a critical balance is obtained. For example, if for any 
reason some of the sodium hydroxide, or more exactly, the hydroxyl 
ions in the solution, are removed from the sphere of action, some of 
the metallic hydroxide will be precipitated from the solution. To 
illustrate, assume for the moment that we have produced a solution of 
metallic hydroxide critically balanced with sodium hydroxide, using 
aluminate as an example. The reaction is given by 

A1(OH) 3 + SNaOH ^ H 3 A1O 3 + SNaOH i> N a3 AlO 3 + 3H 2 O (5) 

A critical point is reached when there is sufficient OH-ion concentra- 
tion to prevent the reaction from reversing. When the sodium hy- 
droxide concentration is increased up to a certain point, the reactioi 
equilibrium is shifted to make available sodium aluminate molecule 
upon which acid aluminate can become peptized to form an agglomei 
tion of molecules or peptized group. The fact that the re-soluti< 
of the aluminum hydroxide occurs rather suddenly, or, in other w< 
the reaction equilibrium is rather critical, indicates that a li 
number of acid aluminate molecules may be peptized on a singl 
molecule of sodium aluminate. Thus, a slight shift in the read 
equilibrium may produce a large change in the solution of the ah 
num hydroxide. 

Consider now what would happen if we were to add a solution 
pure developer to this critically balanced solution. 

The common property of developers is that they have very mu< 
less developing power when in a neutral or non-alkaline solutioi 
The reason for this is the alkali acts as an energizer to ionize the d( 
veloping agent. The simplest example is hydroquinone. 
structure of this component is a benzene ring with two hydroxide 
groups attached at opposite ends. The presence of sodium hydroxide 
in a solution of this developing agent acts to split off the hydrogen 
atoms, leaving an ionized hydroquinone. The hydrogen atoms, thus 
split off, combine with the hydroxide group from the sodium hy- 
droxide to form water. The developing agent is now chemically in a 
condition hi which it is capable of reducing the silver bromide to 
metallic silver. 


- Na+ 

2H 2 (4) 

2Br~) + 2Ag J, (5) 

It is evident from the foregoing that a certain amount of sodium 
hydroxide will be used in the reaction with the developer. Thus, if 
we add a developing agent to a balanced solution of metallic hy- 
droxide, some of the hydroxide will be used up throwing the reaction 
equilibrium out of balance. Less hydroxyl ion concentration re- 
mains than is necessary to hold the metallic hydroxide in solution, so 
a precipitate is formed. This can again be re-dissolved by increasing 
the OH-ion concentration beyond the critical point. 

The function of the balanced solution may be stated to act as a 
buffer to hold the alkalinity of the solution relatively low, while at 
the same time the ionization balance of the organic reduction agent is 
shifted in such a manner to produce increased energy of the developer. 
The reaction of the organic reducing agent with the exposed silver 
bromide in the film produces a shift in ion concentration through a 
long chain of reactions. When the developer is oxidized by silver 
bromide, Br~ ions are released. The reaction of a developer with 
silver bromide is an oxidation reduction reaction, wherein the silver 
is reduced to the metallic state and the developer is oxidized to a less 
negative state. Aluminate ions exist from the reaction. 

A1(OH) 3 + SNaOH => Na 3 AlO 3 + 3H 2 O ^ 3Na+ + AlOa + 3H 2 O (6) 

The only thing that keeps the aluminum hydroxide in solution is an 
excess of sodium hydroxide. The presence of Br~ ions would shift 
the balance of the reaction slightly to the left. Aluminum hydroxide 
is now better able to combine with the gelatin of the film to harden it 
as in any tanning action. Thus, the film in the near neighborhood of 
the developed image undergoes a pronounced hardening which is 
evidenced by a distinct lack of swelling. Furthermore, a general 



[J. S. M. P. E. 

hardening effect over the entire surface of the film is obtained. The 
aluminum hydroxide precipitated during the course of development 
is not entirely absorbed by the gelatin of the film. The major portion 
of it forms a sludge-like precipitate which behaves as a clarifying 
agent, carrying down with it undesirable impurities such as oxidized 
developer and certain other undesirable reaction products which 
would interfere with the course of further development. This pre- 
cipitate can easily be filtered from the solution. 

The presence of the aluminate in a developing solution becomes use- 
ful beyond the point of stabilization of the developer energy. It 
behaves as a very effective inhibitor of swelling of the gelatin film. 
It was found possible to use a developer containing aluminate in 
temperatures well above 110F. To be sure, such extreme condi- 
tions would not be encountered in commercial laboratory practice. 
However, it is seen by this that certain new characteristics arise 
which may prove useful for certain specific purposes. For example, 
the incorporation of the aluminate principle in developers for ama- 
teur use would have a certain advantage inasmuch as processing 
could be carried out in tropical climates or extremely warm weather 
without fear that the emulsion would melt. 


Comparative tests and measurements were made showing the 
relative exhaustion characteristics of the aluminate developer and a 
representative positive developer. The formulas for the two de- 
velopers used in the comparative tests are as follows : 


Sodium sulfite 
Sodium Hydroxide 
Potassium Alum 
Potassium Bromide 


Sodium Sulfite 


Sodium Carbonate 

Potassium Bromide 


RCA Original 




1000 cc 

50 grams 

15 grams 

30 grams 

40 grams 

7 grams 

1000 cc 

RCA Replenisher 


1000 cc 

Sept., 1939] 



In the tests made between the RCA and the Al-Q developer it was the 
object to have as nearly an equivalent amount of developing agent 
as possible in each of the two developers. The first test was carried 
out by exhausting the two developers without replenishment. The 
procedure was as follows: H & D measurements were made with a 
given amount of fresh developer after which a quantity of fully ex- 
posed film was fully developed in the solution to be tested. Further 
H & D measurements were made during the course of this exhaustion. 
Fig. 1 shows the result of this comparative test. In this graph the 


IZOO 24OO 3feOO 


FIG. 1. Comparative test of Al-Q and RCA developers. 

density for a given step on the developed 116 strip is plotted against 
the footage of developed film. Although the test was carried out 
with fully exposed, fully developed film, the measurements of footage 
have been converted to footage of normally exposed, normally de- 
veloped film where the average density of the developed film is taken 
to be 1.25. To facilitate measurements of exhaustion rate it was 
found necessary to use fully exposed film and develop it fully. Sub- 
sequent tests were made with film exposed in a printer and developed 
to an average density of 1.25. The results of these measurements 
were in accordance with the first test. 



[J. S. M. P. E. 

The problem of replenishment has long been one of interest to the 
motion picture industry. It was found possible by use of the alumi- 
nate formulas to replenish partially exhausted developer with the 
original developer formula. Evidently the stability of the developer 
is sufficiently great to obviate the necessity for special booster formu- 
las. Fig. 2 shows the test carried out on the RCA developer where 
replenishment was made at the rate of 200 cc. of replenisher for each 
10 feet of fully exposed, fully developed film. 

A lib strip was developed in the fresh developer, after which 10 

ZO 10*. SO SO*. 4O 4OR. &O &OB. 


FIG. 2. Replenishment test of RCA developer. 

feet of fully exposed film was developed fully in the developer. A 
second lib strip was then put through the developer; 200 cc. of the 
developer was discarded, being replaced by replenisher; and a third 
lib strip was developed. This procedure was repeated until 50 feet 
of fully exposed film had been developed. As can be noted, there is a 
slight general falling of! of the developed density during the course of 
this process of exhaustion. 

A similar procedure was carried out in testing the Al-Q-101 de- 
veloper, although in this case the replenishing rate was 125 cc for 
50 feet of fully exposed, fully developed film, as compared to 200 cc 

Sept., 1939] 



for 10 feet for the RCA developer. It was found that this replenish- 
ing rate was sufficient to maintain the energy constant. The results 
of this test are shown in Fig. 3. 

Fig. 4 shows the characteristic curves obtained during exhaustion 
of the two developers under test. From these curves, speed and 
gamma have been plotted against footage of developed film per gal- 
lon, as shown in Fig. 5. 

As shown by the data obtained in the tests, the new developer ap- 
parently has a life such that constant developing conditions can be 


100 too a \90 


FIG. 3. Replenishment test of Al-Q-101 developer. 

maintained by replenishing with the original formula only for that 
amount of developer which is carried away by the film in the emul- 
sion. In fact, if our understanding is correct, considerably more de- 
veloper is carried away by the film in commercial laboratories than 
would be necessary for maintaining constant activity of the solution. 
Some laboratories do not maintain a closed system for developer 
circulation, in which case the developer is exposed to a considerable 
amount of air. Due to the high energy of the new developer the 
oxidation is greater than with present developers. Tests made by 
bubbling air through small quantities of the developer show that the 



[J. S. M. P. E. 

oxidation properties are not too serious. Fig. 6 shows the results of 
this test. 

A test made in Camden showed that while a deep tank of the de- 
veloper solution (30 gallons with 100 square-inches of surface ex- 
posed) would keep indefinitely, a shallow tank of 10 gallons with 4 
square-feet of exposed surface would oxidize within a period of four 
to six weeks. Actual storage tanks, however, are so designed as to 
present a comparatively small free surface, and in a reasonably active 
laboratory exhaustion, and not oxidation while standing, would be 
the determining factor with respect to deterioration. 





I. BO 




Tos i_oc E 


FIG. 4. Exhaustion characteristics of RCA and Al-Q developers. 

Early in the course of investigations of the new developer the 
question arose whether any special handling consideration was neces- 
sary in respect to neutralization of the acidity in the hypo. The 
Al-Q developer is considerably more alkaline than ordinary develop- j 
ers. It should be expected that due to the carry-over of alkali to the \ 
hypo the acidity of the hypo would be neutralized somewhat more' 
rapidly than in standard developers, thus possibly requiring some! 
acid replenishment. A test was carried out to satisfy this question j 

Sept., 1939] 



where the hardness of the film, fixed in fresh and partially contami- 
nated hypo, was measured. The degree of hardness was determined 
by the procedure specified by the Eastman Kodak Company. Quot- 
ing from p. 109 of Motion Picture Laboratory Practice: 

The degree of hardening produced by an acid fixer and hardening bath may be 
determined as follows: Develop, fix (for normal time used in practice), and wash 
a piece of motion picture film using the solutions to be tested. Place in a beaker 
of water and heat the water so that the temperature rises about 10 F each minute. 
Agitate the film gently at intervals by lifting it out and reimmersing it in the water 



FIG. 5. Exhaustion characteristics of RCA and Al-Q developer. 

and note the temperature at which the gelatin begins to melt or flow away from 
the film base. This melting point is a measure of the degree of hardening, and for 
normal processing conditions should be between 130 and 170 F (54 and 76 C). 

Results of the hardness measurements on film developed in the RCA 
developer and the Al-Q-101 developer are shown :'n Fig. 7. 

It should be noticed that there is one important difference in the 
hardness of films processed in the two developers. Film processed in 
the RCA developer has the same degree of hardness for both the clear 
gelatin and gelatin with the silver image. Film processed in the 
Al-Q-101 developer has a hardness characteristic for the clear gelatin 



[J. S. M. P. E. 

similar to the RCA, but the hardness of the gelatin in the neighbor- 
hood of the silver image is extremely high. In the few tests which 
were made the silver image was found to have a relative hardness 
above 200F. The reason for this can be traced back to the previous 
statements regarding the chemical equilibrium between sodium 
aluminate and aluminum hydroxide present in the solution. In the 
near neighborhood of the development reaction there is a shift in 
alkalinity in such a way as to liberate aluminum hydroxide which 
combines with the gelatin of the film to harden it. In machine 


FIG. 6. Oxidation characteristics. 

processing of motion picture film, developer is continually carrier 
over through the rinse water into the acid hardening fixing bath 
The presence of alkaline developer in the film emulsion acts to neu 
tralize the acid in the acid hardener. As shown in Fig. 7 the effec 
of contamination of the fixing bath by developer solution is more pro 
nounced with the Al-Q developer than with the RCA developer, al 
though this contamination may reach 3 l / z to 4 per cent by volum 
before the hardening effect is appreciably diminished. An optimur 
rinsing time was found to be slightly less than one minute. 

Sept., 1939] 




It was found that variations in developer characteristics with 
respect to temperature were diminished 50 per cent in the Al-Q de- 
veloper. Fig. 8 shows the result of a test plotting temperature and 
degrees Fahrenheit against developed density and gamma for a 
typical sound-recording emulsion. 

Another interesting characteristic of the Al-Q developer, with re- 
gard to temperatures, is the fact that development may be carried 
out at temperatures as high as 115F. As was stated before, such 







* fl IZ 16 2O Z4 


FIG. 7. Results of hardness measurements on film developed in RCA and Al-Q- 

101 developers. 

extreme conditions of temperature would not be encountered in labo- 
ratory practice, but the fact that this can be done serves to illustrate 
that there is a pronounced inhibition of swelling of the gelatin in the 
film. The inhibition of swelling obtained along with the hardening 
of the developed image serves to decrease the turbidity in the emul- 
sion, producing improved resolving power. The image portions are 
hardened in such a way as to diminish the tendency for the silver 
grains to shift their positions. 

For certain purposes (for example, for variable-area sound re- 
cording) high contrast is desirable. Using certain modifications of 



[J. S. M. P. E. 

the Al-Q formula it is possible to develop a film to a contrast beyond 
the characteristic gamma-infinity of the emulsion itself. Generally 
speaking, the definition of the practical gamma-infinity for a particu- 
lar emulsion and developer is the gamma obtained under prolonged 
development. If development of the low densities is inhibited, the 
effective gamma is increased. It was found possible to develop films 
in such a way that up to a given time of development toe densities 
were inhibited so as to produce a greater gamma than would be ob- 
tained upon full development. A time-gamma curve in this case 
would appear as shown in Fig. 9. 

FIG. 8. Variations of developer characteristics with temperature. 

Although extreme conditions can be obtained by suitable modified 
tion of the developer, normal characteristics may be had for soun< 
negatives and positive prints. Fig. 10 shows curves plotted for ; 
sound recording negative material and ordinary print positive film. 

These two characteristic curves were made at the same time a 
test recordings analyzing the cross-modulation characteristics ofl 
tained in the new developer. The method of cross-modulation fo 
use in determining optimum print density for variable- width sooH 
recording has been discussed previously in the JOURNAL. Briefl}i 

Sept., 1939] 



the method is to record a 9000-cycle signal modulated with 400 cycles. 
Image spread in the film can be detected as a distortion of the 9000- 
cycle frequency. If the recorded track has no distortion it will be 
impossible to detect any 400-cycle signal when the film is played in a 
film phonograph. Cross-modulation or image spread will produce a 
400-cycle note, the amplitude of which can be measured and plotted 
as densitometric level. In Fig. 10 are shown cross-modulation curves 
for the new developer Al-Q-101. 

The RCA curve in Fig. 11 shows the characteristic cross-modulation 
for a series of prints from a negative of density 2.1. In this case the 
films were developed in standard developers, modifications of D-16. 


FIG. 9. Time-gamma curves showing increase of 
effective gamma.. 

Here we have print density plotted against densitometric level in 
decibels. If a distortion level of 36 db is taken to be suitable sound 
quality, it will be noted that the range of densities suitable in the 
print lies between 1.6 and 1.82. The density drop from the negative 
density is 0.42. Accordingly, if the negative density is 2.1 the opti- 
mum print density for this developer film combination turns out to be 
1.67. There are two possible improvements on a set-up of this sort. 
For example, if the curve could be broadened out in such a way as to 
give a greater range of suitable print densities, we would thus have a 
greater tolerance available in processing technic. Furthermore, if 
the negative-print density drop could be diminished we could, for the 



U. S. M. P. E. 



Al Q-101 




FIG. 10. Curves of sound negative and print positive film. 


FIG. 11. Cross-modulation characteristics. 


same negative density, have a higher print density which would pro- 
duce somewhat less background noise. In Fig. 11 the curves Al-Q 
show cross-modulation characteristics where the film was developed 
in one of the Al-Q formulas. The negative-print density difference is 
diminished considerably and the curves are broadened so that at a 
densitometric level of 36 db there is considerably greater print 
density tolerance. It was found that slight modifications of the 
developer formula could be made to broaden this density tolerance. 

It should be stated specifically that all the cross-modulation tests 
were made on a special continuous, non-slip, non-synchronous printer 
in order that variations due to printer behavior might be minimized. 
Accordingly, the amount of improvement in cross-modulation char- 
acteristics in commercial work by use of the new developer would 
depend chiefly on printer quality. 


In many of the formulas used in the field it is necessary to dissolve 
the chemicals in a certain specified order. Furthermore, the mixing 
sometimes requires a procedure which is somewhat cumbersome and 
inconvenient. The new developer can be mixed in any fashion what- 
soever and, if desired, the ingredients may be poured as rapidly as 
convenient into a tank of cold water which is being thoroughly agi- 
tated. Chemicals may even be mixed dry and dumped in bodily, 
although, of course, thorough agitation must also be maintained. 
The laboratory tests conducted thus far have indicated many de- 
sirable characteristics. Briefly, they are as follows : 

(1) Long, useful life which, it is hoped, will result in substantial savings to the 
motion picture industry. 

(2) Higher possible contrast and improved resolution for both variable-area 
recordings and release prints. 

(5) Developing conditions with respect to effective emulsion speed, gamma, 
and density hold reasonably constant throughout the life of the developer. 

(4) Temperature effect is reduced by 40 to 50 per cent. 

(5) Density tolerances with modulated high-frequency recording are increased 
by approximately 100 per cent. 

(6) The developer is entirely satisfactory for both sound negatives and pic- 
ture positives. 

(7) The developed image is hardened in the developer. 

(8) The developing time of the formula as at present constituted is of the order 
of two minutes for picture prints and sound positives, and three minutes for vari- 
able-area sound negative. If for any reason somewhat greater optimum develop- 
ing time is warranted, this can be maintained. 

(9) Mixing is simple. 


In closing, recognition should be given to J. O. Baker of the RCA 
Laboratories at Camden for his assistance in the work; to Dr. Solo- 
men of RCA for his counsel ; and to the management of the RCA Re- 
search and Engineering Department for carrying this work through 
its long period of incubation. 


1 CRABTREE, J. I., AND IVES, C. E.: "A Replenishing Solution for a Motion 
Picture Positive Film Developer," /. Soc. Mot. Pict. Eng., XV (Nov., 1930), p. 

2 BAKER, J. O., AND ROBINSON, D. H. : "Modulated High-Frequency Re- 
cording as a Means of Determining Conditions for Optimel Processing," /. Soc. 
Mot. Pict. Eng., XXX (Jan., 1938), p. 3. 


MR. CRABTREE: Have you made tests with the same formula omitting the 
aluminum salt and simply buffering the solution with sodium bisulfite? 

MR. ALBURGER: We have tried that combination. The properties were not 
much superior to those with ordinary carbonate but the characteristics were 

MR. CRABTREE: It is probable that the properties of this developer are 
peculiar to the formula type rather than to any specific effect of the aluminum ions. 
Admittedly, the presence of aluminum ions will tend to tan the gelatin but any 
developer which has sufficient salt content will permit development at 90 F or 

MR. ALBURGER: We have been working on this problem for the last year and 
a half, and have endeavored to cover the situation as completely as possible. We 
have satisfied ourselves, at least, that the problem has been as stated. 



Summary. Improvements in the technic of recording and printing during the 
past few years have made possible the production of 16-mm. sound-films, either by 
optical reduction or by direct recording, having considerably better quality than is 
being obtained in general commercial practice. By the use of a moderate degree of 
equalization in recording, it it practicable to obtain from 16-mm. negatives prints 
giving a flat frequency response to 6000 cycles, with useful response to 7500 cycles, 
when reproduced through a flat amplifying system. Harmonic and envelope dis- 
tortion and speed variations can be kept within acceptable limits for high-quality 
reproduction. The principal remaining defect is background noise. Some general 
agreement upon commercial 16-mm. reproducing system characteristics is needed, 
however, before this improved quality can be made generally available. 

The optical reduction equipment that is in general commercial use 
today for producing 16-mm. sound-films was developed during the 
years 1933, 1934, and 1935. This equipment was described and the 
results obtained were discussed in papers by Baker; 1 Dimmick, Bat- 
sel, and Sachtleben; 2 Kellogg; 3 Batsel and Sachtleben; 4 and Sand- 
vick and Streiffert, 5 in the JOURNAL of this Society. 

All these authors are in close agreement in reporting that the 16- 
mm. prints showed high-frequency losses of the order of 12 decibels 
at 5000 cycles and 16 decibels at 6000 cycles, on the basis of constant 
modulation in the 35-mm. recording. 

Several papers relating to 16-mm. sound have appeared in the 
JOURNAL since the time of the Sandvick and Streiffert paper, but 
none of these has contained direct information on the frequency 
characteristics being obtained from 16-mm. film. The purpose of 
the present communication is to bring the record up to date. 

Since 1935 considerable improvements have been made in 35-mm. 
recording technic, but these have had little effect upon the quality 
of 16-mm. sound-prints commercially available. A good idea of the 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 14, 1939. 

** The Berndt-Maurer Corp., New York, N. Y. 




LT. S. M. P. E. 

quality that is being obtained in practice may be gained by a study 
of the SMPE 16-mm. Sound Test-Film. 6 

Fig. 1 shows the result of measurement on a print of this test-film 
obtained from the New York office of the Society in March, 1939. 
The curve marked A was obtained by running the film on a repro- 
ducer employing a scanning beam 0.0004 inch wide, the measuring 
amplifier having been adjusted to give a flat output when the photo- 
cell was illuminated with constant modulated light at frequencies 
covering the range in which we are interested. The curve marked B 
is obtained by measuring the amplitude of the recorded frequencies 
with a micrometer microscope. The curve marked C is the differ- 
ence between the first two curves. It will be observed that this 


FIG. 1. Characteristics of S.M.P.E. 16-mm. Sound 
Test-Reel. (A) Reproduced characteristic using 

0.0004-inch scanning beam; (B) Measured amplitude; 
(C) Loss characteristic in terms of constant amplitude 

curve repeats rather closely the loss measurements of the authors 
previously cited. Evidently the use of ultraviolet light in recording 
the more recent 35-mm. test-film negative has not materially affected 
the response of the 16-mm. print. 

The amount of equalization shown in curve B of Fig. 1 is greater 
than is commonly used in 35-mm. recording. In practice most 16- 
mm. optically reduced sound-films sound considerably less "bright" 
than would be indicated by curve A of that figure. The practice of 
recording film with special equalization for reduction to 16-mm. is 
rarely employed, though the possibilities are obvious from the results 
shown in Fig. 1 . 

Most of the authors of this series of papers on optical reduction 



printing condemned direct 16-mm. recording as an inferior method. 
This, however, was on the basis of recording image widths in the 
range from 0.0005 inch (Batsel and Sachtleben) to 0.00025 inch 
(Kellogg). The results are very different when the recording image 
is reduced to a width of 0.00015 inch. 

The method of measuring the width of the recording image is im- 
portant if valid comparisons are to be made. A figure for image width 
arrived at by dividing the width of a physical slit by the ratio of re- 
duction of the lens used to image it on the film is not correct unless 
allowance is made for the diffraction pattern and the aberrations of 
the lens. The image widths stated in this paper are arrived at by 






FIG. 2. Characteristics of white-light recording on 
16-mm. positive film. (A) Response with constant- 
amplitude negative, non-slip white-light print; (B) 
Equalization employed in practice; (C) Overall re- 
produced characteristic of non-slip print; (D) Re- 
sponse of optical white-light print from same negative as 
A; (E) Overall reproduced characteristic of optical 
white-light print. 

looking directly at the aerial image formed by the recording optical 
system, using a 4-mm. apochromatic microscope objective of the type 
that is corrected to work without a cover glass. A micrometer ocular 
calibrated in combination with this objective reads with fair repro- 
ducibility to 0.00002 inch. The aberrations and diffraction pattern 
of this apochromat lens are small in comparison with those of the 
image being measured. 

The results of white-light recording on 16-mm. positive film stock 
with an image width of 0.00015 inch are shown in Fig. 2. Curve A 
shows the measured losses of non-slip white-light prints from several 

318 J. A. MAURER [j. s. M. P. E. 

constant-amplitude 16-mm. negatives having densities in the range 
from 1.0 to 1.7, the print densities covering the same range. The 
variation of high-frequency output with negative and print densities 
is too small to be shown on these graphs. 

Curve B of Fig. 2 shows the equalization commonly employed in 
recording 16-mm. sound negatives with white light, while curve C 
shows the resulting overall characteristic. It has been our experience 
that films made according to Curve C are more often criticized as 
"too bright" than as being deficient in high-frequency response in 
comparison with optically reduced prints. The reason for this will be 
apparent when we come to consider the characteristics of commercial 
16-mm. reproducers. 

No curve is given for the response obtained by running the 16-mm. 
sound negative on a reproducer because it has been found that the 
relationship between negative and print is not one that can usefully 
be represented by plotting their individual frequency response curves. 
Thus, at certain densities the print gives a better frequency response 
than the negative from which it was made. It is not the light-modu- 
lating ability but the contrast of the high-frequency portions of the 
negative that is important. For this reason "printer loss" can not 
usefully be expressed in decibels. 

Optical printing of 16-mm. negatives appeared likely to offer the 
advantage of increased effective contrast of the negative, resulting 
from the collimated nature of the light, together with the possibility 
of providing for negative shrinkage without encountering the syn- 
chronization difficulties of the non-slip type of printer. In order to 
find out whether or not these advantages could be realized in prac- 
tice, a printer was built, using a recorder movement at each end of a 
one-to-one optical system made up of the best available microscope 
lenses. White-light prints made with this printer gave I 1 / \ decibels 
higher response at 6000 cycles than non-slip prints from the same 

Curves D and E of Fig. 2 show the effect of optical printing from 
the negatives that gave the non-slip prints represented by Curves A 

As may be seen from Fig. 2, films made by direct 16-mm. recording, 
using standard positive film and white light, and printed by white 
light, are at least as good from the standpoint of frequency response 
as prints made by optical reduction. But this is by no means the best 
that can be done. 



When we have improved optical definition to the point where we 
encounter the law of diminishing returns (and it has been found that 
a recording image narrower than 0.00015, while attainable, gives prac- 
tically no improvement in results) we must perforce turn to methods 
of increasing the resolving power of the film used for recording. 

One of the methods that has proved practicable for securing better 
resolution is the use of ultraviolet light, 7 which confines the image to 
the surface of the emulsion. This method has the disadvantage of 
requiring extremely high light-source temperature if sufficient ex- 
posure is to be obtained. Trials showed, however, that it is not nec- 


50 100 zoo SOO 1000 zaoa 3 + S 7 loooo. 

FIG. 3. Results of high-resolution recording technics. 
(A} Measured response of 16-mm. optical print from 
constant-amplitude 16-mm. negative recorded with 
yellow-dyed stock (Eastman 5504) and Jena BG-12 
filter. Jena UG-3 filter used in printer; (B) Mea- 
sured response of 16-mm. optical print from constant- 
amplitude 16-mm. negative recorded on Eastman 5359 
stock with Jena UG-3 filter. Print through Jena UG-3 

essary to use an actual ultraviolet filter in order to obtain substan- 
tially the same improvement. Any filter that absorbs the band of 
wavelengths from 560 to 450 millimicrons will serve the purpose. 
Such a filter, which is purple or lavender in color, used in combination 
with a fast type of sound-recording film such as Eastman Type 5359, 
requires very little increase of exposure over that required for white- 
light recording with ordinary positive (Type 5301), while it gives a 
response curve only one decibel lower at 6000 cycles than is obtained 
by true ultraviolet recording. A print made optically from a con- 
stant-amplitude 16-mm. negative recorded in this manner, using in 
the printer a filter of the same type as that used in the recorder, shows 

320 J. A. MAURER [j. s. M. P. E. 

a loss of 12 decibels at 6000 cycles and of 16 decibels at 7000 cycles. 
The curve is shown in B of Fig. 3. 

Another method of improving resolution is the use of yellow-dyed 
film stock in combination with a blue filter having a transmission that 
does not overlap the transmission of the yellow dye. This means, in 
general, a filter that absorbs wavelengths greater than 480 millimi- 
crons. The exposure required for this method of recording is as 
great, with films now available, as is required for ultraviolet record- 
ing, but the method has the advantage that the increased light can be 
obtained by redesigning the optical system, without the necessity of 
operating the lamp filament at a high temperature in order to get 
ultraviolet radiation. Furthermore, the results are superior to the 
results of ultraviolet. A negative made by this yellow-dyed-stock- 
plus-blue-filter technic, printed on ordinary positive stock with the 
lavender filter previously referred to in the optical printer, gives a 
print having a loss of only 9*/2 decibels at 6000 cycles and 11V2 deci- 
bels at 7000 cycles. The complete curve is shown by A in Fig. 3. 
A print on the yellow-dyed stock would be still better, but this is not 
practical because pictures can not be printed satisfactorily on 
yellow-dyed stock. 

A third method of improving high-frequency response is the use of 
slow, fine-grained, high-resolution film stocks such as Eastman Type 
1360. Unfortunately this type of stock was not available in 16-mm. 
size at the time these measurements were made, so that no compara- 
tive results can be given. 

Fig. 4 shows an equalization characteristic suitable for use with 
either of the two methods of improving high-frequency response rep- j 
resented in curves A and B of Fig. 3. The two lower curves, B and C\ 
of Fig. 4, show the resulting overall frequency characteristics. Ex- 1 
perience has shown that this amount of equalization, sufficient to give! 
flat reproduction up to 6000 cycles, may be employed without danger 
of causing high-frequency waves to overshoot before those of lower 

Wide frequency ranges in reproduction give satisfactory perform- 
ance only when distortion is kept within suitable limits. Measure- 
ments have been made of the harmonic distortion present in 16-mm. 
prints of 400 and 2000 cycles, produced by both white-light recording 
and by the technics corresponding to Fig. 3. These measurement' 
show that for proper combinations of negative and print densities thf 
total harmonic distortion is less than 5 per cent. The results check 


in general the findings of Sandvick, Hall, and Streiffert, 8 who reported 
in 1923 that the condition for best wave-form (with white-light re- 
cording and printing) was equality of negative and print densities. 
If this condition is satisfied, envelope distortion 9 is also reduced to a 
satisfactory degree. The requirement of accuracy in the control of 
printing is rather exacting, but by no means impossible to meet. 

It is hoped that the work on distortion summarized in the preceding 
paragraph may form the subject of another communication to the 
Society at a future meeting. 

One of the principal difficulties encountered in early attempts to 
build 16-mm. recorders, before the development of optical-reduction 
printing, was the control of speed variations. But, like many of the 
other difficulties of sound-on-film recording, this is mainly a matter of 
mechanical accuracy. The degree of freedom from speed variation 
that has been achieved can best be judged from the reproduction of 
critical musical instruments in the selections that are to be demon- 

The principal remaining defect is background noise. In this re- 
spect prints from direct 16-mm. recordings have a distinct advantage 
over optical reduction prints. The conditions for freedom from 
envelope distortion lead to higher densities in prints from 16-mm 
negatives. Most optical reduction prints have densities in the range 
from 0.9 to 1.2. If they are made denser, they exhibit noticeable 
envelope distortion. Prints from direct recordings, on the other 
hand, may be made as dense as 1.5 with excellent results. The higher 
print densities give noticeable reductions in the noise level of quiet 

At the present time, when reproducing a 7500-cycle frequency 
range, the background noise is about equally divided between film 
noise and photocell hiss. Any development that would make it pos- 
sible to concentrate more light into the scanning beam of the repro- 
ducer would bring a distinct improvement in the hiss level. From 
the standpoint of film noise the need is for finer-grained printing 
stocks. Both these problems will doubtless be solved in the course of 

In its commercial aspects, however, the noise problem is very differ- 
ent. Sixteen-mm. sound projectors in general have been designed to 
reproduce optical-reduction prints. As has been mentioned, these 
prints are generally made from standard 35-mm. recordings, which 
means that the amount of equalization employed is altogether in- 



[J. S. M. P. E. 

sufficient to offset the high-frequency losses. Many of the prints, if 
played on a flat reproducing system, give the impression that the fre- 
quency characteristic is much nearer to Curve C than to Curve A of 
Fig. 1. To offset this loss of high frequencies, most 16-mm. repro- 
ducing amplifiers are peaked at the higher end of their ranges. This 
peak accentuates the background noise, both from the film and from 
the photocell. 

Today it is impossible to employ, commercially, 16-mm. films having 
the improved high-frequency response discussed in this paper. On 
the better reproducers now on the market such films sound objec- 

A / 

FIG. 4. Practical utilization of high-resolution 16- 
mm. recording: (.4) Suitable equalization character- 
istic; (B) Overall response with yellow-dyed nega- 
tive, optical print through Jena UG-3 filter; (C) Over- 
all response with negative recorded through Jena UG-3 
filter, optical print through Jena UG-3 filter; (D) Sug- 
gested reproducer characteristic for most pleasing final 

tionably shrill. In many cases there is considerable attenuation of 
low frequencies, which adds to the unsatisfactory nature of the result. 
If it could be done by general agreement, the author feels that it 
would be desirable to reproduce high-quality 16-mm. films with an 
amplifier characteristic somewhat similar to that shown in D of Fig. 4. 
This procedure would be equivalent to what has been done by the 
adoption of the standard 35-mm. theater reproducing characteristic. 
Noise would be greatly reduced, while the proportion of high fre- 
quencies would still compare satisfactorily with that of 35-mm. 
theater reproduction. Optical-reduction prints having the necessary 
high-frequency content could be produced either by reducing from 


suitably equalized 35-mm. negatives or, more simply, by the use of 
ultraviolet filters in the reduction printing. But unless such a pro- 
gram can be generally agreed upon, higher-quality 16-mm. sound- 
films must be restricted to special applications. 

This, of course, does not mean that these improved methods of re- 
cording are of no immediate value. If they are not used to increase 
the frequency range, they can be employed to increase processing 
latitude or to reduce distortion. In both these directions progress is 


1 BAKER, J. O.: "Sixteen-Mm. Sound on Film," 7. Soc. Mot. Pict. Eng., XXII 
(Feb., 1934), p. 139. 

2 DIMMICK, G. L., BATSEL, C. N., AND SACHTLEBEN, L. T. : "Optical Reduc- 
tion Sound Printing," J. Soc. Mot. Pict. Eng., XXIII (Aug., 1934), p. 108. 

3 KELLOGG, E. W. : "The Development of 16-mm. Sound Motion Pictures," 
/. Soc. Mot. Pict. Eng., XXIV (Jan., 1935), p. 63. 

4 BATSEL, C. N., AND SACHTLEBEN, L. T. : "Some Characteristics of 16-Mm. 
Sound by Optical Reduction and Re-Recording," /. Soc. Mot. Pict. Eng., XXIV 
(Feb., 1935), p. 95. 

5 SAND vi CK, O., AND STREIFFERT, J. G. : "A Continuous Optical Reduction 
Sound Printer," /. Soc. Mot. Pict. Eng., XXV, (Aug., 1935), p. 117. 

6 "S.M.P.E. 16-Mm. Test Films," /. Soc. Mot. Pict. Eng., XXX (June, 1938), 
p. 654. 

7 DIMMICK, G. L.: "Improved Resolution in Sound Recording and Printing 
by the Use of Ultraviolet Light," /. Soc. Mot. Pict. Eng., XXVII (Aug., 1936), p. 

8 SANDVICK, O., HALL, V. C., AND STREIFFERT, J. G.: "Wave-Form Analysis 
of Variable-Width Sound Records," /. Soc. Mot. Pict. Eng., XXI (Oct., 1933), 
p. 323. 

9 BAKER, J. O., AND ROBINSON, D. H.: "Modulated High-Frequency Record- 
ing as a Means of Determining Conditions for Optimal Processing," J. Soc. Mot. 
Pict. Eng., XXX (Jan., 1938), p. 3. 


DR. MILLER: Some years ago we should have thought this an excellent per- 
formance from 35-mm. film. As a matter of fact, from the appearance of these 
16-mm. frequency characteristics, it seems as if we may have to look to our laurels 
on 35-mm. film before long. Will Mr. Maurer give an estimate of the volume 
range that is available in the 16 mm. film we have just heard? 

MR. MAURER: With a reproduced frequency range of 7000 cycles, the back- 
ground noise, as read on an ordinary copper-oxide rectifier type of meter, is 
about 40 decibels below maximum signal. I say "about" 40 decibels because the 
noise varies several decibels with different degrees of cleanliness in processing. 
In practice a volume range of about 30 decibels in the recorded signal may be used 

324 J. A. MAURER [J. s. M. P. E. 

without annoyance from background noise during periods when the modulation 
is at a low level. 

DR. FRAYNE: Will Mr. Maurer please tell us about this special film, and 
whether it is commercially available? 

MR. MAURER: I do not know whether or not it is available in 35-mm. It is a 
yellow-dyed stock, Type 5504, which has been used for some time for reversal 
picture duplication in 16-mm. It is not very suitable for recording because it 
requires quite high exposure, but I hope that before long yellow-dyed film will be 
available that will overcome this difficulty. 

MR. HAWKINS: What amount of noise-reduction was used, in terms of the 
standard 35-mm. noise-reduction? 

MR. MAURER: Very nearly the same practice was followed as is generally 
followed in 35-mm. The width of the biased track is about the same as is found 
on optically reduced 16-mm. prints. 

DR. MILLER: Can you give me any information regarding the optical system 
employed in the original recording? 

MR. MAURER: The optical system forms a line image on the film, which is a 
reduced image of the filament of the recording lamp. The filament is 8 mils in 
diameter, and is reduced to 0.15 mil on the film. 

MR. LAMBERT : Have you made any records available on reversible type film 
suitable for photographing and recording. 

MR. MAURER: I do not have any records here that I can play. Reversible 
film has characteristics that are favorable in one respect and unfavorable in other 
respects. It gives a better frequency characteristic than is obtained by recording 
and printing with a given color of light, but reversible film requires an accurately 
controlled exposure in variable-area recording; otherwise it gives a considerable 
amount of envelope distortion. However, once the exposure has been determined 
for a given type of film it can be maintained in practice within close enough 
limits. The improvement in frequency characteristic is about half as great as is 
obtained by the yellow-dyed stock technic. 

MR. BURCHETT: What results do you get by using that method with Koda- 

MR. MAURER: In direct recording, that is, using a newsreel type of camera, 
very good results are obtained. Kodachrome duplicates are still in the experi- 
mental stage so far as sound is concerned. 

MR. PALMER: Have you done any work hi determining what the maximum 
cut-off would be in recording and printing on Kodachrome? My experience has 
been that it is impossible to print beyond 4000 cycles instead of 6000 or 7000 
cycles, as demonstrated here. 

MR. MAURER: My experience with Kodachrome does not indicate any 
fundamental limit as to the frequency range that can be recorded. The difficulty 
I have experienced is in getting adequate volume in reproduction. 

MR. PALMER: That has been my experience. However, I have noticed a 
definitely lower resolving power of the Kodachrome emulsion as compared with 
reversible black-and-white emulsion we ordinarily use. 

MR. MAURER: That does not run parallel with my experience with Koda- 
chrome. However, I believe that any statements regarding Kodachrome dupli- 


cates at the present time are premature, because it is quite evident that a great 
deal of experimental work is still being done toward improving the process. 

MR. GRIFFIN: I presume that your equipment runs at 36 feet a minute; 
what type of scanning system is being used? Is it the type of scanning system 
such as we call the rotary stabilizer? 

MR. MAURER: The film was run at the standard 16-mm. sound speed of 36 
feet per minute. In referring to the scanning system I take it you mean the 
method of film transport. It is a system that we have developed ourselves, and 
differs rather radically, I believe, from any of those that are in commercial use 
for 35-mm. 



Summary. Fluorescent lamps provide a new tool for solving some of the lighting 
problems encountered in the motion picture industry. A low-pressure glow discharge 
in mercury vapor and an efficient fluorescing powder are the chief features of such 
lamps. The differences between them and the ordinary filament lamps are many 
not only with regard to design but with respect to their operating characteristics. 
These differences are discussed and many characteristics are shown for both the lamps 
and the necessary auxiliary equipment. The daylight quality of light which may be 
produced at high efficiencies and the relative lack of radiant heat make the new lamps 
attractive for several applications in studios. 

The new fluorescent lamps announced to the trade about a year 
ago are receiving more and more attention and offer greater possi- 
bilities for lighting, for decoration, for use with air-conditioning, and 
as a duplication of daylight, than they did at the beginning. They 
are apparently destined to play an important part in the scheme of 
illumination and decoration in many industries as well as in homes, 
shops, etc. The attraction that they have comes somewhat from 
their newness, of course, but is chiefly the result of their high ef- 
ficiencies, their relatively low brightness, their many attractive colors, 
their shape which is so suitable from an architectural standpoint, 
and their ability to product light of daylight quality. Character- 
istics of these fluorescent lamps for low voltage circuits have been 
previously described, 1 ' 2 ' 3 but new information is constantly being 
made available. An approach to the unusual features and a better 
understanding of the characteristics of the new fluorescent lamps may 
be had by a comparison with the well known tungsten-filament lamps. 
Such comparisons are made in the first part of this paper. 

Design. In the ordinary filament lamp the source of light is a 
filament of tungsten wire, either coiled or straight, which is mounted 
on a stem in a bulb, either in vacuum or in an inert gas at nearly 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received May 
15, 1939. 

** General Electric Co., Cleveland, Ohio, and fHollywood, Calif. 


atmospheric pressure. The source of light in a fluorescent lamp is a 
special and carefully controlled synthetically produced powder which 
adheres to the inner bulb surface. Five different powders or "phos- 
phors" are in general use which when used alone give the following 
colors: calcium tungstate blue; magnesium tungstate blue- white; 
zinc silicate green ; zinc beryllium silicate yellow- white ; cadmium 
borate pink.* Theoretically, fluorescent light is caused by the 
movement of electrons of some of the atoms of the powder under the 
influence of ultraviolet radiation. The bulb contains a drop of mer- 
cury, to produce a vapor at low pressure, and argon gas at low pres- 
sure for starting purposes. A tungsten filament becomes incandes- 
cent due to heat from the watts consumed in accordance with the PR 
law, while the fluorescent powder becomes luminous because of its 
ability to transform the short ultraviolet radiation, to which the 
fluorescent powders are particularly sensitive, into visible light. 
This is done by an electric discharge at low current-density between 
the electrodes (the electrodes act alternately as cathodes and anodes, 
depending on the current flow) which are placed in the two ends of the 
bulb (Fig. 1). The cathodes are heated by the discharge during lamp 
operation, and with the selection of proper bulb dimensions, lamps 
low enough in voltage to operate on standard line voltages without 
a step-up are possible. The mercury- vapor must be at low pressure 
for best results, because as the pressure increases there is a shift in 
energy from the X2537 Angstrom line to the longer and less effective 
mercury lines. The low current and vapor-pressure are therefore two 
reasons for not making lamps of the present design in higher-wattage 

The best form for a filament in most cases is one that is concen- 
trated in order to conserve heat, but with fluorescence there is an 
advantage in long lengths. This is due to the fact that there is a 
constant loss at the electrodes, as evidenced by a voltage drop in 
their vicinity, and since the lamp voltage increases with lamp length 
this loss drops to a lower percentage, the greater the lamp length. 
Long tubular sources of light are the result. Four lamps are now 
available, in lengths from 18 to 48 inches, the electrical ratings for 
which are shown in Table I. 

Different colors may be produced in fluorescent lamps at uniformly 

* The chemical names are only partially descriptive of the phosphors which, 
for the most part, are covered by patents and patent applications, as are also cer- 
tain processes for obtaining special color and efficiency characteristics. 



high conversion efficiencies by merely changing the powders on the 
bulb wall, while filaments produce essentially one kind of light and 
changes in color are obtained by the filter method which throws 

Dimensions and Electrical Data 

18-inch T-8 

18-inch T-12 

24-inch T-12 

36-inch T-8 


Lamp Watts 






Diameter (Inches) 






Line Volts 






. 110-125* 

Lamp Amperes 






Lamp Volts 






Rated Life (Hours) 2000 
* Depending on Auxiliary. 





away those colors that are not wanted. Blue, green, gold, pink, red, 
white, and daylight are available in fluorescent lamps, although spe- 
cial color coats for the gold and red are required in order to obtaii 
the desired hues. The lumen output and brightness values are shoi 
in Table II. 

Efficiencies are limited in filament lamps because of the evaporatioi 
and melting of tungsten at the high temperatures, where most ligl 
is given off. Fluorescent lamps are limited by their ability to 

FIG. 1. Internal construction of a fluorescent lamp. 

duce ultraviolet and by the efficiency of conversion of ultraviolet 
into visible. Fig. 2 shows the distribution of energy in the 15-watt 
daylight fluorescent lamp ; for other wattages the proportions remain 
about the same. Skillful lamp design results in about half of the 
energy going directly into the X2537 line, to which the fluorescent 
powders are particularly sensitive. A little less than two per cent 
of the energy is represented in the four visible mercury lines, which 
produce a total of about 45 lumens or three lumens per watt inde- 
pendently of the powders. The rest of the input (along with the 
X2537 conversion loss) is emitted as long infrared radiation or dissi- 

Sept., 1939] 




Lumen Output and Brightness* 
(Approx. Initial Values} 








* Maximum values measured at center of lamp normal to the axis. At 75 de- 
grees from the normal at the center or normal to the axis at the ends the bright- 
ness will be about half the values shown. 

** New Type (3500 K). 

pated by convection and conduction. Part of this unavoidable loss 
keeps the electrodes heated an essential condition for the free emis- 
sion of electrons at practical operating voltages. An average-size 
filament lamp (40-watt) gives 12 lumens of white light per watt 




Lu- Lam- 
mens berts 

450 1800 
530 2100 
310 1260 
900 3600 
300 1200 
380 1500 
45 180 


Lu- Lam- 
mens berts 

450 1200 
530 1400 
310 830 
900 2400 
300 800 
380 1000 
45 120 


Lu- Lam- 
mens berts 

660 1300 
760 1500 
460 870 
1300 2500 
440 850 
540 1050 
60 120 


Lu- Lam- 
mens berts 

1110 2000 
1320 2300 
780 1400 
2250 4000 
750 1350 
930 1650 
120 200 


Lu- Lam- 
mens berts 

1600 1400 
1880 1700 

-20 Waits, I3-X- *+-5 5 W 4 m, 37$*|*5.8 W*its, 39$-+-l 7 Warn, M$->| 

-Total Rdd.6l.on. 7 5 Walts, 50$ *f< 7 5. Warn, 50$ +\ 

13 Walls, 87$- 

FIG. 2. Disposition of total energy delivered to a 15-watt daylight fluores- 
cent lamp. 

while the 15-watt fluorescent (exclusive of loss in auxiliary) gives 
30 lumens per watt, or approximately the same amount of light. 
The luminous efficiencies for the lamps alone vary from 3 lumens per 
watt for the red lamp to 75 lumens per watt for the long green lamp. 
Overall efficiencies are 10 to 25 per cent lower, depending upon the 


G. E. INMAN AND W. H. ROBINSON [j. s. M. P. E. 

type of auxiliary used. Efficiency comparisons with filament lamps 
depend not only upon the size and color of the fluorescent lamp but 
also upon the size, design, and color of the filament lamp. 

Light of daylight quality is produced by combining three fluores- 
cent powders in the same bulb. The spectral curves of the t ree 
overlap and cover the entire visible spectrum. The resultant light is 
close to that of natural daylight both by measurement and observa- 
tion, and is something that has not been produced by the so-called 
daylight filament lamps except when additional highly absorbing 
filters are used (Fig. 3). 







4. . 

_ . 

















\ ' 




- -~~ 












C - D * YLI ^/W,8P^ NT 



t .. 







4000 5000 6000 7000 

3. Distribution curves comparing three daylight sources. 

Lamp Characteristics. A useful by-product of fluorescent lighting 
is a smaller amount of radiated energy with respect to the total watts 
consumed than is encountered with filament lamps. Most of the 
wattage used by filament lamps is given off as infrared radiation. 
This is to be expected because of the characteristics of a hot tungsten 
filament. The fluorescent lamp, however, not dependent upon the 
same principle of light production, does not radiate much heat from 
its relatively cool light-giving substance. The energy that is lost is 
chiefly conducted and convected away from the lamp (Fig. 2), so 
that for equal lumens there is approximately one-fourth as much 
heat radiated from daylight fluorescent lamps as from ordinary low- 
wattage filament lamps. This is a factor in connection with use in 
air-conditioned interiors. 

There is an effect noticeable in the colder climates, due to the tem- 
perature sensitivity of lamps, that should be mentioned and guarded 

Sept., 19391 THE FLUORESCENT LAMP 331 

against. At low temperatures the mercury-vapor inside the fluores- 
cent lamp has a tendency to condense (especially when exposed to 
draft and wind) and reduce the lumen output unless the lamp is 
protected by its fixture or a special shield. The curves shown in 
Fig. 4 indicate the effect produced by various air conditions upon a 
15-watt T-8 lamp with and without a ! 3 /4-inch glass tube, which 
represents about the ideal size enclosure. The draft conditions were 
obtained by placing a 12-inch fan near the lamps. While it is to be 
expected that variations from these specific conditions would result 
in somewhat different curves, these data represent a range that will 
apply to most conditions. For example, an idea of the performance 
of lamps used in an inclosed cavity could be obtained by taking into 
account the cavity size, shape, exposure, and other conditions that 
determine the departure of the installation from the test represented 
here. Effective protection even at 0F is provided by the enclosing 

Every lamp, no matter what kind, when burned in the usual man- 
ner on alternating current has a non-uniform light output due to the 
cyclic variation of current, which is increased at lower frequencies. 
Whether or not this non-uniformity results in an objectionable flicker 
depends upon other things as well as upon the frequency. In fila- 
ment lamps a small wire-size is conducive to variations because of its 
more rapid cooling between half-cycles. In electric discharge lamps, 
where practically no energy is stored up in the light-giving medium, 
the light drops almost to zero along with the current between each 
half-cycle. Fluorescent powder, however, except for the blue-fluores- 
cing variety has a persistence of glow, or phosphorescence, which 
helps to reduce cyclic light fluctuation, and this characteristic is 
affected by the kind of phosphor being used. Ordinarily the slight 
fluctuations from fluorescent lamps are not noticed on 60-cycle cur- 
rent; and with lamps burned on two or more phases or with the use 
of the new two-lamp power-factor-corrected auxiliaries, the fluctua- 
tion in light output is further reduced and is brought to a level 
comparable with that from the ordinary low-wattage filament lamp 
(Fig. 5). 

Auxiliaries. Because of negative volt-ampere characteristics 
every fluorescent lamp requires some sort of ballast, and for the sake 
of starting at a low voltage a simple starting device is attached. 
These are adequately provided by means of a simple reactor and 
switch the latter being momentarily in the closed position, connect- 


G. E. INMAN AND W. H. ROBINSON [j. s. M. P. E. 

ing the cathodes in series with the chbke across the line, for preheat- 
ing purposes; then opening to allow the arc to strike. A very small 
low-cost choke, although it may be adequate for ballasting purposes, 
has a higher loss and does not provide as good regulation as a larger 
choke of higher quality. In practice an automatic thermal type of 
switch has been used with the former, while an automatic magnetic 
switch has been sold with the latter. Both types of switches are 



Shaded areas show probable 
range of opeiatmg effi- 
ciencies of nciosed ' and 
exposed lamps. 

* 15-watt T-8 lamp enclosed in 1 'i inch 3(454 tube. 
A correction has been applied to compensate for 
tube absorption, so the data represent actual lamp 
output Draft pfovided by f*n near lamp 


30 40 50 SO 70 80 

Air Tempefatars. Degrees Fahrenheit 


FIG. 4. Effect of various surrounding conditions on the light 
output of 18-inch T-8 fluorescent lamps. 

somewhat fragile and, consequently, the complete auxiliaries must 
be handled with care. To be satisfactory an auxiliary must not only 
deliver the correct wattage to the lamp but always preheat the cath- 
odes to the proper value and then provide a wave-shape during opera- 
tion that does not show an excessive peak current. Power-factors 
usually run between 50 and 60 per cent, but condensers of the proper 
size to fit in wiring channels are available that largely correct them. 
A new two-lamp auxiliary promises to provide practically unity 

Sept., 1939] 



With changes in line voltage, fluorescent lamp characteristics are 
decidedly different from those of filament lamps. As the line voltage 
increases, the lumen output increases, the wattage increases, and the 
current increases, but the lamp voltage and the efficiency decrease. 
Greater changes take place with a small type of choke : a one-per cent 
increase in line volts may mean a four or five-per cent increase in 
current in this case, while with a large choke only a two-per cent 
change in current may be found. The lumens, although affected 

Operated with Reactors on 115V-60 Cycle-A.C. 

% Deviation from 

Lamp Mean 
Blue 95 

Green 20 

Pink 20 

Gold 30 

Red 10 



Daylight (2 lamps 

operated out of phase) 
15- Watt Filament 

% Deviation from 






FIG. 5. Cyclic flicker chart for 15-watt fluorescent lamps. 

to a lesser degree show the same trend, the change with a small choke 
being about the same as for a filament lamp. Figs. 6(^4) and (B) 
show the characteristic curves. Too low a line voltage (less than 
85 to 90 per cent of normal) may result in uncertain starting and may 
cause a decrease in lamp life, as will operation at excessively high 
voltage. It is desirable to use line voltages close to the center values 
of 118, 208, and 236 volts for the various listed auxiliaries, in order 
to obtain best results. 

Fluorescent lamps may be operated on direct current by combining 
a suitable resistance with the regular auxiliary. Losses in this case 
are higher resulting in lower overall efficiencies. 

As some of the applications of the fluorescent lamps of particular 







95 100 105 






FIG. 6. Changes in fluorescent lamp characteristics at 
varying line voltages. 

04) Using a small, low-cost auxiliary. 

Using a high-quality auxiliary. 

Sept., 1939] 



interest in the motion picture field may be mentioned the use of 
daylight fluorescent lamps for make-up work ; sketching of costumes 
and sets; for mixing paint in colors called for in sketches; matte 
painting; for illumination in close-up motion picture shots; and for 
general interior lighting of offices, commissaries, and the like. These 
are but a few of the many possible applications, and doubtless many 
more will arise with experience and familiarity with the lamps. 

FIG. 7. 

Use of fluorescent lamps for close-up in scene from Warner Bros. 
The Old Maid. 

Fig. 7 illustrates an application of fluorescent lighting for close-ups, 
showing Tony Gaudio of Warner Bros. First National Studio, photo- 
graphing a scene from The Old Maid, starring Bette Davis, Miriam 
Hopkins, and Cissie Loftus, who is shielded by the fluorescent lamp 
in the foreground. A Y-l straw gelatin was used to filter the daylight 
fluorescent lamps to give the best rendition of Miss Davis' blue eyes. 


1 INMAN, G. E., AND THAYER, R. N.: "Low- Voltage Fluorescent Lamps," 
Electrical Engineering, 57 (1938), p. 245. 

2 INMAN, G. E.: "Characteristics of Fluorescent Lamps," Trans. Ilium. Eng. 
Soc., 34 (1939), p. 65. 

3 THAYER, R. N., AND BARNES, B. T.: "The Basis for High Efficiency in 
Fluorescent Lamps," J .Opt. Soc. Amer., 29 (1939), p. 131. 


Fellow Members of the Society of Motion Picture Engineers: 
By tradition and custom, the President of this Society is charged 
with the obligation of making an address to you upon the convening 
of each of your Semi- Annual Meetings. Formerly, you were regaled, 
enlightened, and entertained with a second Presidential address at 
the Semi- Annual Banquet. But with the progress of time, the insis- 
tent demand for this second speech has diminished so that now, it 
would take a very brave and irrepressible President even to attempt 
a banquet oration. 

For the next few minutes, however, I will briefly review the aims 
of our Society and its accomplishments since we last convened here in 
Hollywood. But before proceeding with that portion of this address, 
may I pause to express to our officers, members, and other good 
friends here, our delight at being again in this Capitol of picture 
entertainment. It is always a pleasure to forsake the rigors of our 
Eastern winter and bask in the sunshine of lovely Southern Cali- 
fornia, but it becomes doubly delightful when one is here surrounded 
by the forces of practical creation in that field in which we all labor, 
and where we can commune and exchange ideas and experiences with 
those of you who are responsible for so much that is technically fine 
and artistically beautiful in the motion picture of today. 

The Society of Motion Picture Engineers was organized by Mr. 
C. Francis Jenkins in October, 1916. It grew out of an urgent need 
for just such an organization whose functions were not being and 
could not be performed by any other organization then in existence. 
The subsequent growth and usefulness of our Society is fairly common 
knowledge to all of you. From a membership of nine persons at the 
start, we have grown to a membership of 1319 as of December 31, 

The Society of Motion Picture Engineers is exactly what its name 

* Presented at the 1939 Spring Meeting at Hollywood, Calif., April 16, 1939. 


implies. It is not a commercial or trade association, it has no axe to 
grind, no work to perform other than to serve as a clearing house for 
orderly progress in the development of the science and art of motion 
picture production, processing, and projection. It is recognized by 
the United States Bureau of Standards and by all other leading en- 
gineering and scientific societies as the engineering center of our in- 
dustry. It numbers among its members the leading technicians in 
the motion picture field whether they be in Hollywood, or New York, 
in production or projection, on university staffs or associated with 
commercial organizations, in research work or as executives of the 
companies they represent. This society exists to serve our industry, 
not to compel it. It believes in whole-hearted and generous co- 
operation among our various members and with sister societies. In 
its principal task of standardization, it keeps in mind at all times the 
axiomatic definition of a law as being "the statement of the recur- 
rence of a phenomenon." The standard follows approved practice 
it does not precede it. 

How is it that a society so large as the SMPE can work efficiently? 
The answer is "through committees." The organization of our 
society is the result of practical experimentation, just as the design 
of a piece of technical apparatus is perfected through experimenta- 
tion. Our vice-presidents all have specific functions to perform and 
each is responsible for the work of the comittees they appoint and 
who serve under their guidance. The excellent arrangements for 
this particular Convention are the result of the planning of our 
Executive Vice- President, Mr. Nathan Levinson, and his various 
local committees. The general framework of this, as with other 
conventions for a number of years past, has been the work of our 
genial Convention Vice-President, Mr. William C. Kunzmann. 
The various technical committees, such as the "Laboratory Practice," 
the "Preservation of Film" and the "Projection Practice" committees 
carry on this work of coordination and standardization under our 
Engineering Vice-President, Dr. L. A. Jones. The planning and 
solicitation of papers for this, and many other Conventions of our 
society in the past, and the supervision of the editing of our monthly 
Journal has been under the untiring guidance of our Editorial Vice- 
President, Mr. John I. Crab tree. And last but by no means least 
from a practical point of view, the building of membership and super- 
vision of finances comes under the watchful eye of our Financial 
Vice-President, Mr. Arthur S. Dickinson. Your Secretary and 

338 E. A. WlLLIFORD [J. S. M. P. E. 

Treasurer perform the usual duties of those offices while your Presi- 
dent supplies the general planning and, we hope, the urge and inspira- 
tion to accomplishment and harmony. 

Our Society, like our government, functions under a Constitution. 
What we can do for the industry or for any individual company or 
person is limited by the powers granted us in this Constitution, and 
the By-Laws which supplement it. Provision has been wisely made, 
however, for amendment to this charter and the number of such 
changes which have been made from time to time by action of the 
society, has enabled us to keep abreast of changing times and chang- 
ing needs. 

No organization, no matter how perfect on paper, is any better 
than the men who occupy the responsible positions in it. New men 
come along to fill vacancies as they are open and some of the old- 
timers are reflected to their former offices because no new man has 
shown up who has the ability, the time, and the willingness to take 
this job. The more active members of our Society are constantly 
looking for new blood to fill places as officers, members of the Board 
of Governors, committee chairmen, and even members of committees 
themselves. It is my belief that in the near future, no one will be 
permitted to succeed himself in office indefinitely, thereby compelling 
new men to come in to take their places. 

With this word picture of our Society in action, let us review for a 
moment the progress which has been made during the past year in 
the motion picture industry. Not since 1931, have so many nor so 
radical innovations in film emulsions been introduced to the world. 
First came films with increased speed and finer grain as compared 
with the former emulsions available. Then came a special, fine- 
grained film with twice the former speed for background projection. 
And finally, for black-and-white photography, there was introduced 
a film with four times the speed of last year's speediest emulsions. 
For color photography a similar high-speed emulsion is available. 

The introduction of these faster films has made possible an im- 
provement in the quality of the photographic image through stopping 
down the lens opening at present levels of lighting. Or, if money 
savings seem more important, the amount of light used can be re- 
duced while using present lens openings. Or perhaps a better bal- 
ance can be arrived at by using some of each of these methods. 

And while on the subject of set lighting, there has continued a 
movement toward smaller units of higher light output made possible 


by improved incandescent bulbs and silent, steady-burning, high- 
intensity arcs. 

Great progress has been made in the art and science of background 
projection. A committee of the Research Council of the Academy 
has developed recommended standards for projectors, lamps, car- 
bons, lenses, and other necessary equipment which has greatly ad- 
vanced the precision and availability of this medium. Triple- 
projector units have been used successfully in superimposing three 
images from keyed prints with satisfactory registration on the back- 
ground screen and greatly increased volume of light. 

This past year has seen much greater use of light and exposure- 
meters on sets. While these have not fully solved some of the prob- 
lems inherent in the use of the faster films and in making color pic- 
tures, they have been of considerable assistance and have pointed 
the way toward further development of such meters. 

In the theater field, new projection machines and new sound sys- 
tems have improved the quality of exhibition. A greater use of the 
high-intensity arc, particularly of the non-rotating positive carbon 
type, has added to the enjoyment of theater patrons, especially when 
color pictures are shown. 

During 1938, the use of 16-mm film was greatly expanded. Some 
producers are making feature pictures available on 16-mm acetate 
film for theatrical as well as for non-theatrical use. New 16-mm 
projectors and cameras with sound have been developed and placed 
on the market. A small, compact carbon arc projection lamp has 
been developed in conjunction with 16-mm projectors giving several 
times the light and definition previously available for 16-mm film. 

Several new 8-mm cameras and projectors have appeared on the 
market during the past year. One of these is adaptable for 8-mm, 
9.5-mm, or 16-mm film, while another is capable of showing either 
8-mm or 16-mm pictures. 

Last, but far from least in the amount of interest engendered, is 
the commercial introduction of television. While direct telecasting 
of sports and news events and perhaps a few dramatic productions 
seem in the offing, by far the greatest use of television will probably 
be in the telecasting of sound-films. For this purpose, both inter- 
mittent and non-intermittent projectors have been used successfully. 
In England, recently, a prize fight was televised directly from the 
ringside to two theaters. Two systems of telecasting and reception 
were used, one employing a cathode-ray tube for projection onto the 

340 E. A. WlLLIFORD 

theater screen and the other using a carbon arc lamp for projecting 
the televised image. Both screens were small but the images were 
reported as fairly distinct and flickerless. The reporter who wit- 
nessed both experiments stated however, that much was yet to be done 
before televised theater pictures would be satisfactory to the patrons. 
During the next few days, you will see and hear demonstrated 
some of the improved materials, apparatus, and methods referred to 
in this brief comment. We hope you will attend as many of our 
sessions as your time permits and that you will participate freely in 
the discussion of the various papers. This is your Society and you 
will benefit from it in proportion as you give of your time and effort. 
In closing, may I wish for all of you, much pleasure and profit during 
the balance of our convention. 

E. A. W. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photo static copies may be obtained from the Library of Congress, Washington, D. C. t 
or from the New York Public Library, N. Y. Micro copies of articles in maga- 
zines that are available may be obtained from the Bibliofilm Service, Department of 
Agriculture, Washington, D. C. 

British Journal of Photography 

86 (June 9, 1939), No. 4127 

Foveal Circle Scan, A Basically New Method of Moving 
Picture Presentation (pp. 362-363) 

86 (June 23, 1939), No. 4129 
Progress in Colour (p. 387) 


19 (May, 1939), No. 5 
New High Fidelity Receiver (pp. 14-15, 35-36) 

Television Lighting, Pt. I (pp. 17-19) 
Films in Television (pp. 28, 30, 32-34) 

19 (June, 1939), No. 6 
Sound Motion Picture Films in Television, Pt. II (pp. 

17, 27, 30) 
The Fundamentals of Television Engineering, Pt. Ill 

(pp. 18, 20, 24) 

Educational Screen 

18 (June, 1939), No. 6 
Motion Pictures Not for Theatres, Pt. 10 (pp. 191-194, 


International Photographer 

11 (June, 1939), No. 5 

Direct Color Still Methods Compared (pp. 5-6) 
Projection Symposium, Pt. 7 (pp. 7-9) 
New Shift Modernizes B & H Camera (pp. 16, 18) 

Kinematograph Weekly 

268 (June 8, 1939), No. 1677 
Television by Scophony's Newest Model (p. 31) 


L. G. PACENT and 







268 (June 1, 1939), No. 16?6 
Title Cards by Typewriting Process (p. 36) 
A Puzzle in Photometry, Solar Mirror Increases Actinic 
Light (pp. 36, 35) 

Philips Technical Review 

4 (April, 1939), No. 4 

A Simple Apparatus for Sound Recording (pp. 106-113) K. DE BOER and 




Officers and Committees in Charge 

E. A. WILLIFORD, President 

S. K. WOLF, Past-President 

W. C. KUNZMANN, Convention Vice-President 

J. I. CRABTREE, Editorial Vice-P resident 

D. E. HYNDMAN, Chairman, Atlantic Coast Section 
J. HABER, Chairman, Publicity Committee 

S. HARRIS, Chairman, Papers Committee 

H. GRIFFIN, Chairman, Convention Projection 

E. R. GEIB, Chairman, Membership Committee 

Reception and Local Arrangements 

D. E. HYNDMAN, Chairman 







Registration and Information 

W. C. KUNZMANN, Chairman 


Hotel and Transportation 

J. FRANK, JR., Chairman 

C. Ross P. D. RIES 



J. HABER, Chairman 



P. A. McGuiRE 

P. A. McGuiRE 


344 1939 FALL CONVENTION [j. s. M. P. E. 

Convention Projection 

H. GRIFFIN, Chairman 






Officers and members of Projectionists Local 306, IATSE 

Banquet and Dance 

A. N. GOLDSMITH, Chairman 





Ladies' Reception Committee 

MRS. O. F. NEU, Hostess 






Headquarters. The headquarters of the Convention will be the Hotel Pennsyl- 
vania, where excellent accommodations have been assured, and a reception suite 
will be provided for the Ladies' Committee. 

Reservations. Early in September room reservation cards will be mailed to 
members of the Society. These cards should be returned as promptly as possible 
in order to be assured of satisfactory accommodations. The great influx of visi- 
tors to New York, because of the New York World's Fan-, makes it necessary to 
act promptly. 

Hotel rates. Special per diem rates have been guaranteed by the Hotel Penn- 
sylvania to SMPE delegates and their guests. These rates, European plan, will 
be as follows: 

Room for one person $ 3 . 50 to $ 8. 00 

Room for two persons, double bed $ 5.00 to $ 8.00 

Room for two persons, twin beds $ 6.00 to $10.00 

Parlor suites: living room, bedroom, $12.00, $14.00, and 
and bath for one or two persons $15.00 

Parking. Parking accommodations will be available to those who motor to 
the Convention at the Hotel Fire Proof Garage, at the rate of $1.25 for 24 
hours, and $1.00 for 12 hours, including pick-up and delivery at the door of the 

Sept., 1939] 1939 FALL CONVENTION 345 

Registration. The registration desk will be located at the entrance of the 
Banquet Room on the ballroom floor where the technical sessions will be held. 
All members and guests attending the Convention are expected to register and 
receive their badges and identification cards required for admission to all the 
sessions of the Convention, as well as to several de luxe motion picture theaters in 
the vicinity of the Hotel. 

Technical Sessions 

The technical sessions of the Convention will be held in the Banquet Room on 
the ballroom floor of the Hotel Pennsylvania. The Papers Committee plans to 
have a very attractive program of papers and presentations, the details of which 
will be published in a later issue of the JOURNAL. 

Luncheon and Banquet 

The usual informal get-together luncheon will be held in the Grand Ballroom of 
the Hotel on Monday, October 16th. The Semi-Annual Banquet and Dance will be 
held in the Grand Ballroom of the Hotel Pennsylvania on Wednesday, October 
18th, at 8:30 P. M. At the banquet the annual presentation of the SMPE Prog- 
ress Medal and the Journal Award will be made, and the officers-elect for 1940 
will be introduced. 


Motion Pictures. At the time of registering, passes will be issued to the dele- 
gates of the Convention admitting them to several de luxe motion picture theaters 
in the vicinity of the Hotel. The names of the theaters will be announced later. 

Golf. Golfing privileges at country clubs in the New York area may be ar- 
ranged at the Convention headquarters. In the Lobby of the Hotel Pennsylvania 
will be a General Information Desk where information may be obtained regard- 
ing transportation to various points of interest. 

Miscellaneous. Many entertainment attractions are available in New York to 
the out-of-town visitor, information concerning which may be obtained at the 
General Information Desk in the Lobby of the Hotel. Other details of the enter- 
tainment program of the Convention will be announced in a later issue of the 

Ladies' Program 

A specially attractive program for the ladies attending the Convention is being 
arranged by Mrs. O. F. Neu, Hostess, and the Ladies' Committee. A suite will 
be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. Further details will be published in a succeeding 
issue of the JOURNAL. 

New York World's Fair 

Members are urged to take advantage of the opportunity of combining the 
Society's Convention and the New York World's Fair on a single trip. Informa- 
tion on special round-trip railroad rates may be obtained at local railroad ticket 


offices. Trains directly to the Fair may be taken from the Pennsylvania Station, 
opposite the Hotel: time, 10 minutes; fare, lOjf. Among the exhibits at the 
Fair are a great many technical features of interest to motion picture engineers. 

Points of Interest 

Headquarters and branch offices of practically all the important firms engaged 
in producing, processing, and exhibiting motion pictures and in manufacturing 
equipment therefor, are located in metropolitan New York. Although no special 
trips or tours have been arranged to any of these plants, the Convention provides 
opportunity for delegates to visit those establishments to which they have entree. 
Among the points of interest to the general sightseer in New York may be listed 
the following: 

Metropolitan Museum of Art. Fifth Ave. at 82nd St.; open 10 A. M. to 5 P. M. 
One of the finest museums in the world, embracing practically all the arts. 

New York Museum of Science and Industry. RCA Building, Rockefeller Cen- 
ter; 10 A. M. to 5 P. M. Exhibits illustrate the development of basic industries, 
arranged in divisions under the headings food, industries, clothing, transportation, 
communications, etc. 

Hayden Planetarium. Central Park West at 77th St. Performances at 11 A. M., 
2 P. M., 3 P. M., 4 P. M., 8 P. M., and 9 P. M. Each presentation lasts about 
45 minutes and is accompanied by a lecture on astronomy. 

Rockefeller Center. 49th to 51st Sts., between 5th and 6th Aves. A group of 
buildings including Radio City Music Hall, the Center Theater, the RCA Build- 
ing, and the headquarters of the National Broadcasting Company, in addition to 
other interesting general and architectural features. 

Empire State Building. The tallest building in the world, 102 stories or 1250 
feet high. Fifth Ave. at 34th St. A visit to the tower at the top of the building 
affords a magnificent view of the entire metropolitan area. 

Greenwich Village. New York's Bohemia; a study in contrasts. Here are 
located artists and artisans, some of the finest homes and apartments, and some 
of the poorest tenements. 

Foreign Districts. Certain sections of the city are inhabited by large groups of 
foreign-born peoples. There is the Spanish section, north of Central Park; the 
Italian district near Greenwich Village; Harlem, practically a city in itself, num- 
bering 300,000 negroes; Chinatown, in downtown Manhattan; the Ghetto, the j 
Jewish district; and several other such sections. 

Miscellaneous. Many other points of interest might be cited, but space permits 
only mentioning their names. Directions for visiting these places may be ob- j 
tained at the Convention registration desk: Pennsylvania Station, Madison 
Square, Union Square, City Hall, Aquarium and Bowling Green, Battery Park, 
Washington Square, Riverside Drive, Park Avenue, Fifth Avenue shopping dis- j 
trict, Grand Central Station, Bronx Zoo, St. Patrick's Cathedral, St. Paul's | 
Chapel, Cathedral of St. John the Divine, Trinity Church, Little Church Around 
the Corner, Wall St. and the financial district, Museum of Natural History, \ 
Columbia University, New York University, George Washington Bridge, Brook- 
lyn Bridge, Triborough Bridge, Statue of Liberty, American Museum of Natural 
History, Central Park, Metropolitan Museum of Art, and Holland Tunnel. 



At the meeting of the Board of Governors, held at the Hotel Pennsylvania, 
New York, N. Y., on July 13th, the following nominations for officers and gover- 
nors for 1940 were made: 

Engineering Vice-P resident D. E. HYNDMAN 

Financial Vice-P resident A. S. DICKINSON 

Secretary J. FRANK, JR. 

Treasurer R. O. STROCK 

Governors H. GRIFFIN 


Nominating ballots have been mailed to the voting membership of the Society, 
and announcement of the results will be made on the first day of the Fall Con- 
vention at New York on October 16th. 

The Engineering Vice-President, Financial Vice-President, and the Governors 
are elected for two-year terms; the Secretary and Treasurer for one-year terms. 
Of the six nominees for Governors, three will be elected. The elected officers and 
governors will assume office on January 1, 1940. 


At the meeting of the Board of Governors on July 13th, the following amend- 
ments to the By-Laws were proposed for submission to the Society at the Fall 
Convention. These proposed amendments are published herein in accordance 
with the requirements of By-Law XI, specifying the method of amending the 
By-Laws, and will be acted upon at the Business Session of the Fall Convention 
scheduled for Monday, October 16th. 



Sec. 1. All committees, except as otherwise specified, shall be appointed by 
the President. 

Sec. 2. All committees shall be appointed to act for the term served by the 
officer who shall appoint the committees, unless their appointment is sooner 
terminated by the appointing officer. 

Sec. 3. Chairman of the committees shall not be eligible to serve in such ca- 
pacity for more than two consecutive terms. 



It is intended that this By-Law, if and when adopted, will be known as By-Law 
IV, the present By-Law IV and all subsequent By-Laws being re-numbered ac- 



(In the following proposed amendment, the new portions are printed in Italics, 
the unchanged portions in Roman type.) 

Sec. 1. (a) All officers and five governors shall be elected to their respective 
offices by a majority of ballots cast by the Active, Fellow, and Honorary members 
in the following manner : 

Not less than three months prior to the annual fall convention, the Board of Governors 
shall nominate for each vacancy several suitable candidates. Nominations shall first 
be presented by a Nominating Committee appointed by the President, consisting of 
nine members, including a chairman. The committee will be made up of two Past 
Presidents, three members of the Board of Governors not up for election, and four 
other Active, Fellow, or Honorary members, not currently an officer or Governor of 
the Society. Nominations shall be made by three-quarters affirmative vote of the total 
Nominating Committee. Such nominations shall be final unless any nominee is re- 
jected by a three-quarters vote of the Board of Governors present and voting. 

The secretary shall then notify these candidates of their nomination, in order of 
nomination, and request their consent to run for office. From the list of accept- 
ances, not more than two names for each vacancy shall be selected by the Board of 
Governors and placed on a letter ballot. A blank space shall also be provided on 
this letter ballot under each office, in which space the names of any Active, Fellow, 
or Honorary members other than those suggested by the Board of Governors may 
be voted for. The balloting shall then take place. 

The ballot shall be enclosed in a blank envelope which is enclosed in an outer 
envelope bearing the secretary's address and a space for the member's name and 
address. One of these shall be mailed to each Active, Fellow, and Honorary 
member of the Society, not less than forty days in advance of the annual fall con- 

The voter shall then indicate on the ballot one choice for each office, seal the 
ballot in the blank envelope, place this in the envelope addressed to the secretary, 
sign his name and address on the latter, and mail it in accordance with the in- 
structions printed on the ballot. No marks of any kind except those above pre- 
scribed shall be placed upon the ballots or envelopes. 

The sealed envelope shall be delivered by the secretary to a committee of tell- 
ers appointed by the president at the annual fall convention. This committee 
shall then examine the return envelopes, open and count the ballots, and announce 
the results of the election. 

The newly elected officers and governors of the general Society shall take office 
on the January 1st following their election. 

(6) The first group of vice-presidents, viz., the executive vice-president, engi- 
neering vice-president, editorial vice-president, financial vice-president, conven- 
tion vice-president, and a fifth governor, shall be nominated by the Board of 
Governors at its first meeting after the ratification of the corresponding provisions 
of the Constitution; and the membership shall vote on the candidates in accord- 


ance with the procedure prescribed in these By-Laws for regular elections of 
officers so far as these may be applicable. 

The following is the original paragraph which the italicized paragraph given above 
is intended to supersede: 

Not less than three months prior to the annual fall convention, the Board of 
Governors, having invited nominations from the Active, Fellow, and Honorary 
membership by letter form not less than forty days before the Board of Governors' 
meeting, shall nominate for each vacancy several suitable candidates. 


As announced in the previous issues of the JOURNAL, the Fall Convention of the 
Society will be held at the Hotel Pennsylvania, New York, N. Y., October 16th- 
19th, inclusive. Details pertaining to the Convention will be found on page 343. 


The following applicants were admitted by vote of the Board of Governors to 
the Active grade: 


716 N. La Brea, 195 Broadway, 

Los Angeles, Calif. New York, N. Y. 


1199 E. 49th St., 60 Holmes PL, 

Brooklyn, N. Y. Lynbrook, Long Island, N. Y. 


4240 Dundee Dr., 304 S. Cannon Dr., 

Los Angeles, Calif. Beverly Hills, Calif. 


British Acoustics, 10531 Bloomfield St., 

Film House, Wardour St., North Hollywood, Calif. 
London, England. 


These films have been prepared under the supervision of the Projection 
Practice Committee of the Society of Motion Picture Engineers, and are 
designed to be used in theaters, review rooms, exchanges, laboratories, 
factories, and the like for testing the performance of projectors. 

Only complete reels, as described below, are available (no short sections 
or single frequencies). The prices given include shipping charges to all 
points within the United States; shipping charges to other countries are 

35-Mm. Visual Film 

Approximately 500 feet long, consisting of special targets with the aid 
of which travel-ghost, marginal and radial lens aberrations, definition, 
picture jump, and film weave may be detected and corrected. 

Price $37.50 each. 

16-Mm. Sound-Film 

Approximately 400 feet long, consisting of recordings of several speak- 
ing voices, piano, and orchestra; buzz-track; fixed frequencies for focus- 
ing sound optical system; fixed frequencies at constant level, for de- 
termining reproducer characteristics, frequency range, flutter, sound- 
track adjustment, 60- or 96-cycle modulation, etc. 

The recorded frequency range of the voice and music extends to 6000 
cps. ; the constant-amplitude frequencies are in 11 steps from 50 cps. to 
6000 cps. 

Price $25.00 each. 

16-Mm. Visual Film 

An optical reduction of the 35-mm. visual test-film, identical as to 
contents and approximately 225 feet long. 
Price $25.00 each. 







Volume XXXIII October, 1939 


Carbons for Transparency Process Projection in Motion Picture 

Studios D. B. JOY, W. W. LOZIER, AND M. R. NULL 353 

Recent Improvements in Carbons for Motion Picture Studio 

Arc Lighting 


A New Mobile Film- Recording System 

Use of an A-C Polarized Photoelectric Cell for Light- Valve Bias 

Current Determination C. R. DAILY 394 

A Densitometric Method of Checking the Quality of Variable- 
Area Prints C. R. DAILY AND I. M. CHAMBERS 398 

A Direct-Reading Photoelectric Densitometer . . . . D. R. WHITE 403 

Acoustic Condition Factors . . . M. RETTINGER 410 

Controlled Sound Reflection in Review Rooms, Theaters, Etc. 

C. M. MUGLER 421 

The Status of Lens Making in America W. B. RAYTON 426 

Motion Pictures in Education A. SHAPIRO 434 

Screen Color W. C. HARCUS 444 

New Motion Picture Apparatus 

A Light- Weight Sound Recording System 


Current Literature 458 

1939 Fall Convention at New York, N. Y., October 16th-19th, 

Inclusive 461 

Abstracts of Convention Papers 465 





Board of Editors 
J. I. CRABTREE, Chairman 




Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Omce, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1939, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


** President: E. A. WILLIFORD, 30 East 42nd St., New York, N. Y. 
** Past-President: S. K. WOLF, RKO Building, New York, N. Y. 
** Executive Vice-President: N. LEVINSON, Burbank, Calif. 

* Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
** Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 

* Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 
** Convention Vice-President: W. C. KUNZMANN, Box 6087, Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St., New York, N. Y. 

* Treasurer: L. W. DAVEE, 153 Westervelt Ave., Tenafly, N. J. 

** M. C. BATSEL, Front and Market Sts., Camden. N. J. 

* R. E. FARNHAM, Nela Park, Cleveland, Ohio. 

* H. GRIFFIN, 90 Gold St., New York, N. Y. 

* D. E. HYNDMAN, 350 Madison Ave., New York, N. Y. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* A. C. HARDY, Massachusetts Institute of Technology, Cambridge, Mass. 

* S. A. LUKES, 6427 Sheridan Rd., Chicago, 111. 

** H. G. TASKER, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1939. 
** Term expires December 31, 1940. 



Summary. Data are given on the amount and distribution of light which can be 
obtained on the transparent screen with regular high-intensity carbons and various 
optical systems. These optical systems include two condenser-type systems and two 
relay-condenser-type systems used with f/2.3 and f/2.0 objective lenses. Values are 
obtained which are in line with the recommendations on Process Projection Equipment 
by the Research Council of the Academy of Motion Picture Arts and Sciences. 

In addition, characteristics of two new carbons developed for this purpose are given. 
These are also evaluated in the relay-condenser system and give the advantage of addi- 
tional, light, or the same light at lower energy input. 


"Process projection" or "background projection" are terms used 
to represent the practice of supplying a background for a motion pic- 
ture set by means of projection from a film through a translucent 
screen placed at the rear of the actual set. This field of cinematog- 
raphy has received only scanty consideration in the literature, but 
persons having first-hand acquaintance with activities in the motion 
picture studios know of the vast amount of work done in the studios 
on process projection, the wonderful results accomplished, and the 
economic and artistic importance of this process to the motion picture 
industry. However, it is only too well realized that the use of the 
process has to a large extent been restricted by the technical short- 
comings of the equipment, and it is only the skill and resourcefulness 
of studio technicians that have made the use of background projec- 
tion as widespread as it is at present. 

The Research Council of the Academy of Motion Picture Arts and 
Sciences appointed hi March, 1938, a committee under the chair- 
manship of Farciot Edouart, composed of motion picture studio 
; experts in the field of process projection and other interested persons, 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
21, 1939. 

** National Carbon Co., Fostoria, Ohio. 


354 JOY, LOZIER, AND NULL [j. s. M. p. E. 

to consider the problems and needs of process projection in its en- 

As a result of a year's intensive work, this Committee has published 
a report of Recommendations on Process Projection Equipment, 1 
released in February, 1939. This report, as stated, "presents for 
the first time, the coordinated viewpoint of the majority of the 
Hollywood studios on this subject and should be of great value to all 
the studios and to the manufacturers of process projection equip- 
ment." This report is certainly worthy of detailed study by all per- 
sons interested in motion picture production and is indicative of the 
high quality of the work of the various committees of the Research 
Council of the Academy. It has been of great benefit in visualizing 
the light-source requirements for process projection and in the de- 
velopment of carbons more suitable for this application. 

This Academy Report describes three possible optical systems for 
use with the carbon arc for process (or rear) projection. It was be- 
lieved that it would be interesting and helpful to measure the light, 
and distribution of light which can be obtained by these systems with 
the carbons recommended for this application, and this paper in- 
cludes such data. 

Furthermore, as a result of studying the particular requirements of 
this process projection, special 11 -mm and 16-mm carbons have been 
developed for this application and are herein described. The 11 -mm 
carbon is of particular importance because it has an intrinsic brilliancy 
far in excess of any carbon hitherto available. As a result of this 
characteristic this carbon at 135 amperes gives with the relay con- 
denser system a light on the projection screen equal to that formerly 
possible only with an arc at 180 amperes on a larger carbon. 


In motion picture theater projection, the film aperture through 
which the light is projected is 0.825 X 0.600 inch and it is desirable 
that there be a falling off in light from the center to the sides of the 
screen. The ratio of the brightness at the sides to that at the centei 
of the screen may vary from 80 to as low as 60 per cent without behi 
unusual in appearance. 

Process or background projection on the other hand uses the mucl 
larger silent camera aperture 0.950 X 0.723 inch in size, and some 
times the sound camera aperture, 0.868 X 0.631 inch. Also the 


desirable distribution of the light on the translucent screen is that 
the brightness at the sides and corners of the screen should be as high 
as, or preferably higher than, at the center. 

The properties of the translucent screens in use today dictate to a 
certain extent this desired distribution of light over the screen. In 
order to obtain sufficiently high transmission of light through the 
screen, its diffusion properties are somewhat inferior to the "ideal" 
diffuser. This gives rise to a "hot spot" seen from the camera at 
i that portion of the screen on the line joining the camera and the 
| projector. In general, this effect combines with the usual decrease in 
illumination at the boundaries of the screen to make a very accentu- 
ated variation in brightness over the screen when viewed from the 
camera position. It is therefore extremely desirable to keep the 
I light falling upon the boundaries of the screen at as high a value as 
possible relatively to that falling upon the center of the screen in 
order to avoid this "hot-spot" effect. 

The combination of the projected background scene with the fore- 
ground action into a single series of images on the motion picture film 
I imposes further restrictions in order to create the desired illusion. 
The intensity of the background must be balanced with that of the 
foreground action. The illumination of the foreground and back- 
ground must remain uniform and free from fluctuations in order to 
prevent disturbing contrasts between the two from becoming notice- 
able in the final motion picture. Finally, in color photography it is 
necessary that the color of the background be balanced with the 
foreground color. 

In process projection, there is not the restricted space of the motion 
picture projection booth limiting the dimensions of the lamp or 
optical system, nor is there the necessity of operating the arc for 
more than 20 minutes of continuous burning. Most rear-projection 
scenes are of only a few minutes' duration. This has placed less limi- 
tation on the design of both carbons and optical systems for process 

There is a demand for more light on the translucent screen and 
this, as stated in the Committee Report, 1 should be combined with 
"absolute steadiness of the projected picture with a minimum of 
light variation on the screen and increased efficiency of the light." 

In addition to the bulletin referred to above there are two earlier 
articles 2 ' 3 on process projection outlining some requirements and 
methods in use. 

356 JOY, LOZIER, AND NULL [j. s. M. p. E. 


This paper will consider only those cases of process projection in 
which motion picture film is projected; still backgrounds obtained from 
stereopticon projection are not directly considered, although some of 
the results discussed below can doubtless be applied here. 

For process projection both the full silent camera aperture and the 
standard sound camera aperture are at times employed; therefore 
in the tests described below both these apertures have been used. 

At least three distinct optical systems have been considered for 
rear projection. These are the reflector- type optical system, the 
condenser-type optical system, and a modified relay-condenser- 
type system developed by the laboratories of the Technicolor Motion 
Picture Corporation. The authors are indebted to the Technicolor 
Company for the details of the latter system. These three types of 
optical systems that have been considered for process projection 
differ among themselves in the degree to which they satisfy the re- 
quirements. In the following, these optical systems are described 
and measurements reported of the luminous flux falling on the screen 
when used with various standard carbons and two experimental 
carbons. The screen illumination was measured at several points on 
the screen using calibrated Weston Photronic cells equipped with 
Viscor filters. The illumination was reduced to convenient values by 
placing calibrated wire screens in the light-beam. 

The actual values of luminous flux falling on the screen are in 
part determined by the speed and design characteristics of the par- 
ticular projection lens employed. The desirable projection lens 
characteristics are described on pp. 12 and 13 of the Academy: 
Bulletin. 1 We have used for these measurements three lenses, all 5 
inches in focal length: the Bausch & Lomb //2.3 Super Cinephor, 
which has been found very satisfactory in design characteristics for 
rear projection work; and two experimental //2.0 lenses Nos. 506 
and 524 loaned to us by the Bausch & Lomb Optical Company, to 
whom we are indebted for their use. Lens No. 524 gives more total 
light and a higher proportion of light at the sides and corners of the 
screen, but this has been done with some sacrifice of the higher 
quality of projection of lens no. 506. 

In the measurements described below values are reported for the 
lumens falling on the translucent screen ; these values were obtained 
with no film shutter. Screen distribution values for the sides and 


corners of the screen are quoted and represent the illumination at the 
sides and corners of the screen relative to that at the center of the 
screen taken as 100. 

Relative heat at the aperture has been measured by placing a suit- 
able device in the plane of the aperture, and all the figures quoted 
below are on a comparable arbitrary basis. The device used con- 
sisted of a thermocouple mounted on a small, blackened silver-plate 
receiver; the combination was calibrated for thermocouple tempera- 
ture vs. total incident radiant energy. 


The basic elements of the ordinary mirror system used in theater 
projection are well known and the optics of this system have been 
fully discussed in the literature 4 ' 5 ' 6 so that it will suffice here to give 
only a summary review. In this system the mirror collects the radia- 
tion from the carbon arc and forms an image of the arc on the film 
aperture; the light passes through the film and is focused on the 
screen by the projection lens. The diameter of the mirror and its 
distance from the film-gate determine how completely the light-beam 
fills the relative aperture of the projection lens; the size of the light- 
source and the magnification factor of the mirror determine how 
adequately the film aperture is covered with light. 

High-intensity reflector type lamps in use in theatrical projection 
today employ an ellipsoidal reflector and are largely of two types: 
first, an angular-trim lamp employing a 9-mm bare positive carbon 
which is rotated; and, second, the more recent horizontal- trim lamps 
employing 6-mm, 7-mm, or 8-mm copper-coated "Suprex" positive 
carbons which are not rotated. The light output would not be 
sufficient for process projection except for very small screens. Also, 
these lamps are designed to illuminate the smaller aperture used in 
theater projection, and will not cover to best advantage the larger 
apertures used in process projection without an increase in the carbon 
size or the mirror magnification. Neither the mirrors nor the lamp 
housing and mechanism are designed to withstand the higher power 
input called for by the use of larger carbons. A current of possibly 
140 to 150 amperes would be necessary with a mirror system to 
take care of transparency projection adequately with the large silent 
camera aperture. Some measurements were made with such a sys- 
tem and total light comparable to that possible with the relay-con- 
denser system was obtained. However, the light at the sides was 



tf. S. M. P. E. 

considerably lower than at the center, and since this was on a purely 
experimental basis, with no assurance that mirrors in a lamp of rea- 
sonable size would stand this usage, these data are not included in the 
present comparisons. 


Condenser-type lamps are used for projection in the larger theaters 
and have been widely used in background projection in the studios. 
The basic features of this system are shown in Fig. 1. The condenser 
lens gathers the radiation at the arc and forms an image of the arc 
crater on the film aperture. In this respect the mirror and condenser 
systems behave alike ; they differ in respect to the magnification and 
collecting angle. In general, the size of the cone of radiation col- 
lected at the arc is smaller with condensers than with mirrors; there- 

, -< 



~~~~~~- ---..J 



. Li r 


r /VJ 




K"" L. 


"-- ( i 

n Pisojecr/o 
'Time Lens 

r//V<? Ri 




CoNOH3/z Are* 

FIG. 1. Diagram of the conventional condenser system. 

fore, in order to fill fast lenses and obtain intense screen light, the 
condenser will have a lower magnification ratio than a mirror having 
a wider collecting angle. This lower magnification of the condenser 
system necessitates the use of a larger carbon than with the mirror 
system. Therefore, we find with condenser systems that the positive 
carbon diameter ranges from 13.6 to 16 mm in diameter. In general, 
the positive carbon is rotated continuously during burning, and the 
negative carbon makes an angle of approximately 40 to 60 degrees 
with the positive carbon. 

Screen light and aperture heat measurements were made on four 
trims of "National" carbons commonly used for rear projection. 
These carbons were burned in the Brenkert Model A lamp. Two sets 
of condensers, listed in Table I, manufactured by the Bausch & 
Lomb Optical Company, were used. Condenser combination A was 
designed a number of years ago, and condenser combination B is a 


set of more recent development. The distances employed with these 
condensers in the tests are given in Table I. 

Measurements of screen light were taken with the three Bausch 
& Lomb projection lenses and two camera apertures previously de- 
scribed, and a summary of the results is given in Table II. 

In this table we have listed the results in such a way that the ob- 
jective lenses can be directly compared under each condenser-lens 
combination, and the condenser-lens combinations compared for 
each carbon. The main grouping is according to the carbons used. 
Each carbon combination was burned at approximately its maximum 
recommended current. 


Distances Used with Condensers for Screen Light Measurements 

Distance from Carbon Distance from Front 

Condenser Combination to Rear Condense* Condenser to Aperture 

Combination A 3 . 75 inches 15 inches 

Rear Lens, B. & L. 41-86-27 

Front Lens, B. & L. 41-86-28 

Combination B 2. 75 inches 15 inches 

Rear Lens, B. & L. 41-86-62 

Front Lens, B. & L. 41-86-63 

The two camera apertures are used since both are employed to 
some extent for process projection. The relative heat at the aperture 
was measured as previously described, and the figures are in arbitrary 
units and give the comparative heat which was obtained at the center 
of the aperture with each carbon and condenser-lens combination. 
Changing the objective lens would not, of course, affect this tempera- 
ture at the aperture. It was felt that the central point of the film 
aperture would be the most desirable place to take these temperature 
readings and that any spill-over at the aperture or other parts of the 
projector mechanism could be taken care of by suitable masks just 
ahead of the film aperture. 

Condenser-lens combination B in every case gives about 30 to 
40 per cent more light at the aperture than condenser-lens combina- 
tion A. This will probably vary somewhat with different lenses 
but is a significant increase. It is at least partly accounted for by 
the fact that the rear lens of combination B is closer to the positive 
carbon and subtends a larger angle of light. Combination A sub- 
tends an angle of 80 degrees, whereas combination B subtends an 

S a 


angle of 95 degrees. Reducing the distance of the rear lens to 2 3 /4 
inches from the positive crater increases the possibility of condenser 
pitting and breakage, and this may be a serious factor, particularly 
with the larger-size carbons at the higher currents. Both these 
condenser systems were positioned to give slightly less than the maxi- 
mum light in order to bring up the distribution of the light at the 
sides and corners of the screen and thus more nearly approximate the 
requirements for rear projection. The total screen lumens for 
the silent camera aperture is, of course, somewhat higher than for the 
sound camera aperture, but this is counterbalanced by the fact that 
the light distribution is poorer on the silent camera aperture than on 
the sound camera aperture. 

The carbons have been listed in the table in the order of the light 
on the screen; in other words, the 13.6-mm high-intensity positive 
carbon at 125 amperes gives the least light and the 13.6-mm super- 
high-intensity positive at 180 amperes gives the most light, with the 
two 16-mm carbons falling between. The 13.6-mm super carbon at 
180 amperes gives with condenser-lens combination B, with the 
//2.3 and //2.0 systems, approximately the values in lumens indi- 
cated as desirable in the Academy Bulletin 1 (p. 22). Likewise, the 
values for the 16-mm super-high-intensity carbon at 195 amperes 
with condenser-lens combination B approximate those values referred 
to above, and the light distribution on the screen is more favorable 
for the 16-mm super carbon. The amount of energy at the arc with 
the 16-mm super carbon is slightly higher than with the 13.6-mm 
carbon. On the other hand, the 13.6-mm carbons burn at a higher 
consumption rate. The other carbons considered burn at lower con- 
sumption rates, and although the light is not as great as that ap- 
parently desirable for many process projection shots, it may be 
sufficient under some conditions for the smaller screens. 

The //2.0 No. 506 lens gave appreciably more screen light and 
about the same distribution of light as the //2.3 Super Cinephor. 
The //2.0 No. 524 lens showed in every case an increase in total 
lumens and higher relative light at the sides and centers than the 
//2.0 No. 506 lens. This is without doubt compensated for in most 
applications by the increased definition, etc., of this No. 506 lens, as 
indicated by the lens manufacturer. 

In order to visualize what the total lumens and the comparative 
screen distribution means in light on the transparency screens, we 
have listed in Table III the same carbon, condenser-lens, and objec- 


tive lens combinations, and have indicated the light in foot-candles 
which would be obtained at the center, sides, and corners of a 14-foot 
screen using the sound camera aperture and the silent camera aper- 
ture. Of course, for the same 14-foot picture the transparency screen 
would have to be farther away for the sound camera aperture than 
for the silent camera aperture. The results listed in Table III are 
significant to one who is accustomed to measuring the light on the 
transparency screen with a foot-candle meter. For example, the 
13.6-mm super-high-intensity carbon at 180 amperes with condenser- 
lens combination B and the //2.0 No. 506 lens will give on a 14-foot 
screen with the silent camera aperture 134 foot-candles at the center, 
96 foot-candles at the sides, and 73 foot-candles at the corners; 
whereas the 13.6-mm high-intensity positive carbon at 125 amperes 
with this same condenser-lens and objective lens combination will 
give on this same 14-foot screen only 103 foot-candles at the center, 
72 foot-candles at the sides, and 56 foot-candles at the corners. 

Compared with the light at the center, the light at the corners and 
sides of the screen is somewhat lower than what is apparently de- 
sirable for process projection with the present transparency screen 
and methods used. This ratio of the light at the sides and the 
corners to that at the center could be improved by changing the posi- 
tion of the condenser-lenses with respect to the carbons, but this 
would entail a considerable loss of total light on the transparency 


An optical system which promises to have particular merit for 
background projection is a modification of the relay condenser sys- 
tem. Relay condenser systems were patented by A. Koehler of the 
Zeiss Works about 1915 7 and Roger Hill in 1927. 8 The former has 
found application in photomicrography and microprojection, where 
it is called "Koehler illumination." 9 Some attempts have been made 
to use this type of optical system for motion picture projection and 
some results obtained with it have been described in the litera- 
ture. 7 ' 10 ' 11 - 12 A modification of this system has been suggested by 
the Technicolor Motion Picture Corporation and is shown in Fig. 2. 

This system was designed for a speed of //1. 6, and differs from the 
conventional condenser system in that the crater 'of the arc is not 
focused on the film aperture P 2 but rather in the vicinity just to the 
left of the lens L 3 . The plane PI, which is intermediate between the 



[J. S. M. P. E 

arc crater and the condenser-lenses, is focused by means of L\, L z , 
and the relay lens LZ, on the film aperture P 2 - In this manner the 
coverage of the film aperture is much less dependent upon the distri- 
bution of the brightness across the arc crater and its size than with 
the conventional condenser system shown in Fig. 1. Here the varia- 
tion of illumination across the film aperture follows more closely that 
across the plane PI than that across the carbon crater. A second 
image of the arc is formed in the vicinity of the projection lens L 4 and 
the size of this image, relative to the dimensions of the projection 
lens, determines how well this lens is filled with light. The size of 
the light-source, therefore, determines the size of the light-beam at the 
projection lens, and hence can affect the total projected light but has 
very little effect on the screen distribution. 




FIG. 2. Diagram of the relay condenser system developed by the Technicolor 
Motion Picture Corporation. 

Previous forms of the relay condenser system imaged a plane at 
the condensers LI and L^ on the film aperture and suffered the dis- 
advantage that any imperfection or noticeable details in or on the 
condensers could be noticeable on the projected screen image. Also, 
it has been the practice in some cases to place an auxiliary aperture 
lens over the film aperture; the same criticism is applicable here as is 
expressed in the preceding sentence. 

In the experimental set-up, lenses LI and L 2 were the same as 
the combination A of Table I, with the single exception that the 
surface of LI nearer the arc was changed from the customary cylin- 
drical surface to a spherical surface of the same curvature. This 
eliminates the oval image of the carbon which would be formed with 
the cylindrical-surfaced lens and which would be of no particular 
value here. L 3 was a 5V4-inch diameter Bausch & Lomb PM25 lens. 


The spacing of the various parts of the optical system is shown in 
Fig. 2. After the lens 3 was set to focus the plane PI on the 
film aperture, and the lens L was focused on the projection screen, it 
was found that the lens Z, 3 , the film aperture PZ, and the lens L 4 could 
all be moved back and forth to obtain a spot of light which covered 
the film aperture. In this manner, the size of the spot was adjusted 
to cover the two sizes of film aperture which were used. It was also 
found desirable to adjust this spacing slightly with the different pro- 
jection lenses to obtain the most satisfactory screen light and screen 

In measuring the relative heat at the aperture, it was found desir- 
able to place a mask in the light-beam at P 3 . This prevented light 
and heat which would not enter the projection lens from passing 
through the film aperture. This procedure could be used to ad- 
vantage in process projection practice where an //2.0 objective lens 
is used, because in several cases, particularly with large-size carbons, 
it has reduced the heat at the center of the aperture 50 per cent, with 
no reduction in light. Five standard carbon trims were measured 
in this relay condenser system using the same lamp, three projection 
lenses, and two aperture sizes already described. In addition, data 
were obtained on two experimental carbons which are the result of 
recent research work and which show outstanding characteristics 
with this relay system for transparency projection. 

Measurements of screen light, screen distribution, and relative 
heat at the film aperture for these regular and experimental carbons 
are given in Table IV. 

The five regular trims of carbons are grouped as in the case of 
Table II, in the order of the total lumens on the translucent screen. 
Each carbon is burned at approximately its maximum current. The 
values for the//2.3 Super Cinephor and//2.0 No. 506 lens are given 
for both the silent and sound aperture, and in some cases the values 
for the//2.0 No. 524 lens also are included. 

The relative heat at the aperture is on the same arbitrary basis as 
in Table II. The total screen light obtained with carbons used in 
this relay condenser system is noticeably greater than that obtained 
on the regular condenser system combination A, and of about the 
same magnitude as with the condenser system combination B. This 
is again accounted for to some extent by the fact that the rear con- 
denser on this relay condenser system is at a distance of 3.15 inches 
from the arc and subtends an angle of light from the crater of 90 

SS 8 



il" Carbo 

% Screen 

78 : 




B O 




U | 




-8 O O 



^ 00 OS 

















S c 
fej ij 






| & co ^ 

S O- o c^i 

"^ co p p 




degrees. This angle of light is somewhat greater than that of the 
condenser system A but smaller than that of the condenser system B. 
This distance, as stated before, was obtained experimentally, and 
appeared to be the most advantageous one from the standpoint of 
total light and distribution for the particular lenses used in this 
system. Even though the total light on the screen is essentially the 
same in amount for this relay condenser system as for the regular 
condenser system combination B, the distribution of the light on the 
screen is considerably more favorable for the relay condenser system 
than for the regular condenser system. In other words, it is desirable 
in process projection or rear projection that the sides and corners of 
the screen have the maximum amount of light compared with the 
center. It should be noted that in many instances in this table, 
the light at the sides of the screen with the sound camera aperture, 
is as high as or higher than at the center of the screen, and even with 
the silent camera aperture this condition is approached. 

With this relay condenser system the 13.6-mm super and the 16- 
mm super again approximated the conditions mentioned in the Re- 
search Council's report; 1 that is, gave more than 12,000 lumens 
with the//2.3 system and more than 16,000 lumens for the//2.0 sys- 
tem. The 16-mm super-high-intensity carbon has a more favorable 
distribution than the 13.6-mm super carbon, the current is slightly 
higher, and the consumption is considerably lower. 

In considering the application and use of carbons for this relay 
condenser system, it was realized that it would be very desirable to 
increase the total amount of light available and to cut down, if pos- 
sible, the amount of energy at the arc and the temperature at the film 
aperture. A research program was therefore undertaken to achieve 
these objectives. As a result, we present two carbons known as the 
16-mm experimental positive and the 11-mm experimental positive, 
which offer considerable improvement in these respects. These 
carbons were designed particularly with the relay condenser system 
in view, and the data on these carbons are presented in the last two 
sections of Table IV. 

It is seen from these data that the 16-mm experimental positive 
carbon gives more than 19,000 lumens with the //2.0 No. 506 lens 
with reasonably good distribution. This is more light than we have 
obtained with any of the other carbons on the same basis. With the 
//2.0 No. 524 lens this value is increased to 22,000 lumens. This 
carbon was burned at 225 amperes and 74 arc volts and has a con- 





sumption rate of 22 inches per hour. This is a very high energy 
input, and it is questionable whether it is practicable to position the 
condenser three inches or less from the arc under these conditions. 

For this reason we feel that the 11 -mm experimental carbon is of 
greater interest, giving at 135 amperes light equal in amount and dis- 
tribution to that produced by the 13.6-mm super-high-intensity car- 





.20 J5 .10 .OS jOS .10 .15 JO 


FIG. 3. Comparison of intrinsic brilliancy across 
the crater face of the "National" 13.6-mm super- 
high-intensity carbon at 180 amperes and the 11-mm 
experimental carbon at 135 amperes. 

bon at 180 amperes. This is well above the minimum desired by the 
Research Council Committee; 1 and furthermore, this 11-mm experi- 
mental carbon gives slightly less heat at the film aperture than the 
13.6-mm super-high-intensity carbon. This differential in heat is 
approximately 15 per cent and is of considerable importance in this 
application. The carbon consumption for this 11-mm experimental 
carbon is 38 inches per hour compared with 28 inches per hour for the 
13.6-mm super carbon. 

3s 1 


en - y n 

^j Tfi 1C 


co co ID co 

o o 

WH 2 ** o 
~- SU 


^ txj ^ | 


.1 i<3 
I pa 







S : 

^ - 



e 2*si 

S ^ (N 

g CO 

00 S CD 

!> 00 


i s 








o * 





<*o ^ 5 



S 3|Js 



O O 

^ to ,_3 

^ CD CO 

co L, '"I 

rH t* 

g S ^ 

"g 00 O 

<N" ^ o 

^" CD" -^ i-T 

rjn CD 



^ > 












I 4 


5 la I 

2 g 



(M O 





S r " NrS 

2* Si 





^ ^ ^c 


^ 056 s - U2 
-S rvi 

"s 1 ~ 

>* "~ 

42 " 








5 Q ! 




*^. H-^ $ 




IT o | 

<o O O 

s 8 

o o 

fcL^ ^ fe ^ 

^ CO 

i* ^O 

c^ c^ 

3 ||| 

2 CD" oo" 


T ( CO ^^ O5 

!-<" CO" 


g ^ 















: . 








^ jz 

| J 


2 g 

OH a 

. ^ 

K; S 


3 o 


hi a 


2 ^ CD 

^ ^ a3 

o ^ .i fc 

CD * 


-S 2 

I I 




^ fc 

b S 



* o a! 
S n 

S?3So 1^ lSo| 

<J ^ -H O -H 


t^- Os CO Oi ^^ 

CD "H 
CO 00 



00 i-i CO = (M (M 
1C l> 00 -< CD l> 



The reason for this marked superiority in light on the translucent 
screen for the 11-mm carbon is evident in Fig. 3 which shows the in- 
trinsic brilliancy distribution curves for the two carbons. The 
experimental carbon gives the heretofore unattained figure of 1200 
candles/sq. mm for the intrinsic brilliancy at the center of the crater, 
or about 300 candles/sq. mm (or 33 per cent) more than the 13.6-mm 
super carbon. It is because of this very high brilliancy that this 
carbon is able to furnish such intense light on the translucent screen at 
a lower energy input than other carbons of lower intrinsic brilliancy. 
Such a carbon can be run with less danger to the condensers and less 
likelihood of burning the film than the other carbons mentioned in 
this table burning at 180 amperes or above, and still give the same 
amount of light. 


Comparison of Total Screen Light and Foot- Candle Readings on 14- Foot Screen 

13.6-Mm Super H.I. Carbon at 180 Amperes; //2.0 No. 506 Lens; Silent Camera 


Optical System 

Screen Lumens 
No Film 

Foot-Candles on 14 
Center Sides 

Foot Screen 

Condenser Comb. A 





Condenser Comb. B 





Modified Relay Condenser Using 

Lenses* from Comb. A 





Modified Relay Condenser Using 

Lenses from Comb. B 





* Rear surface of rear lens spherical instead of cylindrical. 

The results in Table IV are retabulated in Table V on the basis of 
the foot-candles which would be obtained on a 14-foot screen with this 
relay condenser system, and this table is directly comparable with 
Table III, for the regular condenser system. It can be seen by com- 
paring Tables III and V that the relay condenser system gives much 
more light at the sides and corners, where it is most useful, than either 
of the condenser systems. For example, the 13.6-mm super-high- 
intensity carbon at 180 amperes gives approximately 17 per cent 
more light at the sides and corners of the screen with the //2.0 
No. 506 lens than with the condenser combination B, where the rear 
condenser is closer to the arc and therefore in greater danger of being 
cracked. The relay system on the same basis gives 50 per cent more 
light on the sides and 70 per cent more light at the corners than the 

372 JOY, LOZIER, AND NULL [j. s. M. P. E. 

condenser combination A, where the rear element is slightly farther 
from the arc than the relay condenser system. 

As previously stated, the condensers used in this relay system 
are the same as the condensers in combination A except that the rear 
surface of the rear condenser has been ground spherical. It seems 
pertinent therefore to know what would result if the condensers of 
combination B were used in the relay condenser system. We have 
included in this paper the results obtained with three carbon combina- 
tions using this special relay system. It was found desirable to move 
the rear condenser to a distance of only 2 3 / 4 inches from the arc. 
This is as close as the lamp mechanism will allow, and the same dis- 
tance as when these lenses were used in the regular condenser system. 
This closer proximity of the condenser to the arc materially increases 
danger from breakage and pitting. However, the results are very 
interesting and are shown in Table VI. 

The total light is slightly less than with the other relay system, but 
the light at the sides and corners, even with these high-quality lenses, 
is noticeably greater than at the center and therefore of the quality so 
much desired for transparency projection. Also for this reason, the 
foot-candles values at the corners of the transparent screen are much 
higher for carbons used with this modified relay system than with the 
other systems. This is illustrated by Table VII, which compares 
the foot-candle values at the center, sides, and corners of the 14-foot 
screen for a given carbon such as the 13.6-mm super-high-intensity 
carbon used in the four systems in Tables III, V, and VIb. It is 
evident that the two relay condenser systems give greater light at 
the sides and corners, and a much more favorable light distribution. 

It is hoped that these data will be of value to the motion picture 
studio industry. Work on this subject is being continued and we 
expect to be able to report additional results in the not too distant 


1 "Recommendations on Process Projection Equipment," Bulletin, the Re- 
search Council, Academy of Motion Picture Arts and Sciences (Feb. 3, 1939); 
/. Soc. Mot. Pict. Eng., XXXII (June, 1939), p. 589. 

2 POPOVICI, G. G. : "Background Projection for Process Cinematography," 
/. Soc. Mot. Pict. Eng., XXIV (Feb., 1935), p. 102. 

3 POPOVICI, G. G. : "Recent Developments in Background Projection," /. 
Soc. Mot. Pict. Eng., XXX (May, 1938), p. 535. 


4 HARDY, A. C. : "The Distribution of Light in Optical Systems," /. Franklin 
Inst., 208 (Dec., 1929), p. 773. 

6 HARDY, A. C. : "The Optics of Motion Picture Projectors," /. Soc. Mot. Pict. 
Eng., XIV (March, 1930), p. 309. 

6 COOK, A. A. : "A Review of Projector and Screen Characteristics, and Their 
Effects upon Screen Brightness," J. Soc. Mot. Pict. Eng., XXVI (May, 1936), p. 

7 KELLNER, H. : "Can the Efficiency of the Present Condensing Systems Be 
Increased?" Trans. Soc. Mot. Pict. Eng., XVII (1923), p. 133. 

8 U. S. Pat. No. 1,630,616 (May 31, 1927). 

9 "The Principles of Optics," Hardy and Perrin, p. 508, McGraw-Hill Book 
Company (New York) 1932. 

10 KELLNER, H. : "Results Obtained with Relay Condensing System," Trans. 
Soc. Mot. Pict. Eng., XVIII (1924), p. 143. 

11 TOWNSEND, L. M. : "An Improved Condenser System for Motion Picture 
Projection," Trans. Soc. Mot. Pict. Eng., XI (1927), p. 512. 

12 HILL, R.: Trans. Soc. Mot. Pict. Eng., XX (1924), p. 88. 



Summary. Recent improvements in carbons for broadside lamps and spotlamps 
are described. New negative carbons, designed for use with spotlamps result in more 
quietly burning arcs. 

New carbons have been designed for broadside lamps giving quieter burning and 
steadier light. The light is of the same spectral quality as that of the spotlamp-filter 
combinations used for Technicolor photography. Curves of spectral energy distribu- 
tion of light received on the set and records illustrating the improved steadiness and 
quieter burning are shown. 


The requirements which an artificial light-source must fulfill to 
be successfully used for the illumination of motion picture studios 
have been outlined in various articles already published. 1>2>3 These 
requirements have been modified from time to time by technical 
developments in the motion picture industry; for example, the advent 
of sound pictures emphasized the requirement of quietness and the 
more recent expansion of Technicolor has brought its demands for a 
suitable color-balance in the lighting. 

The carbon arc plays a prominent role in supplying both types of 
studio lighting in widespread use today; i. e., in furnishing general 
overall illumination and the intense "modelling" light. New lamp 
designs have been developed 4 ' 5 - 6 to utilize in a very efficient manner 
the high brilliancy and desirable color characteristics of the carbon 
arc. These improvements in lamps have given a more precise 
control over the burning of the arc, resulting in a steadier light and 
minimum noise due to lamp mechanism. 

It is the purpose of this paper to record recent developments in 
carbons for broadside lamps and spotlamps which give additional 
gains in (1) steadier light, (2) quieter burning, and' (3) better color- 
balance for Technicolor photography. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
17, 1939. 

** National Carbon Company, Fostoria, Ohio. 




Spectral Quality of Light. The spectral quality of the light has 

in determined by measuring the spectral energy distribution. 
For visual purposes of comparison the trichromatic coefficients 7 ' 8 
have been calculated and the color-temperature determined accord- 
ing to standardized procedures. 9 

Light Steadiness. The steadiness of the light has been measured 
by a Weston photronic cell connected to a General Electric photo- 
electric recorder to obtain a continuous graphic record of the light- 

Measurement of Sound-Level of the Arc. In order to obtain a 
laboratory measurement of the sound-level associated with the arc, 
a crystal microphone was connected to the input of a resistance- 
coupled amplifier, the output of which was coupled through a trans- 
former to a General Electric photoelectric recorder connected to 
record alternating current. Tests were run late at night when ex- 
traneous noise was at a minimum, and the gain of the amplifier was 
kept at such a level that ground-noises were not a disturbing factor. 
In order better to register the arc noise the microphone was placed 
close to the lamp. The amplifier gain was maintained constant 
throughout a series of measurements and ground-level was taken 
frequently during the series with the arc shut off and without moving 
the microphone. Power-source noise was eliminated by choke coils 
and condensers appropriately connected. This gave a permanent 
record of the comparative quietness of arcs with various experimental 
carbons. It is probably a less significant record than can be ob- 
tained at the motion picture studios where sound-stages and extensive 
recording equipment are available. 


The improvement in spotlamp mechanisms as reported in the 
JOURNAL by Richardson 5 has reduced the sound from the mechanical 
operation of the arc to a very low value. Suitable choke coils and 
condensers have successfully eliminated the generator hum. How- 
ever, there was still sufficient noise from this arc to give concern when 
used very close to the microphone or when many units were used on 
very large sets. 

Ways of reducing this noise are being studied by the Committee 
on Set Equipment Noise Conditions of the Research Council of the 
Academy of Motion Picture Arts and Sciences. They have observed 

376 JOY, LOZIER, AND ZAVESKY [j. s. M. p. E. 

a significant reduction in noise in lamp units lined with sound- 
absorbing material and have recommended the use of such material 
in these spotlamps. 

It is difficult for one not intimately connected with motion picture 
production to visualize the low sound-level with which we are con- 
cerned. However, after conducting tests at night in our own labora- 
tory and observing tests conducted by this Committee in sound- 
proof studios, it was evident that some of this residual noise was 
intimately connected with the arc itself. Research work indicated 

16-Mm H.I. Studio Positive, 13.6-Mm H.I. Positive, 

150 Amperes, 68 Volts 115 Amperes, 57 Volts 

Vie* V/M.P. /' Vie'M.P. 

Ground Orotip Studio Ground Orotip Studio Ground 

Noise Negative Negative Noise Negative Negative Noise 



FIG. 1. Effect of new type negative carbons in reducing noise-level. 

that changes in the negative carbon would noticeably reduce this 

From this research work new negative carbons have been de- 
veloped for use with the 13.6-mm H.I. W.F. and 16-mm H.I. 
studio positive carbons used respectively in the Type 90 and Type 
170 studio spotlamps. These new negative carbons, which are 
known as C.C. M.P. Studio negative carbons, are slightly larger 
than the "Orotip" negative carbons which they are intended to replace; 
a 7 /i6-inch diameter C.C. M.P. replaces the "Orotip" 3 /s-inch negative 
carbon and a V2-inch diameter C.C. M.P. replaces the "Orotip" 
7 / 1 e-inch negative carbon. 


These new negative carbons known as "Motion Picture Studio 
Negatives" have been submitted to the special sound committee 
of the Research Council to evaluate in terms of relative noise-reduc- 
tion under studio set conditions; such tests will be much more 
valuable than measurements made under our laboratory conditions. 
However, these latter measurements do give an idea of the magnitude 
of the reduction and should therefore be of some interest although 
they should be considered as merely preliminary to the Committee's 
measurements under the studio conditions. 

The method used in our tests has been described earlier in this 
paper. Sixteen-mm H.I. Studio positive carbons were operated at 
150 amperes using consecutively the "Orotip" 7 /i 6 -inch negative car- 
bon and the new 1 /2-inch C.C. M.P. Studio negative carbons; like- 
wise 13.6-mm H.I. positive carbons were operated at 115 amperes 
with the "Orotip" 3 / 8 -inch and the new 7 /ie-mch C.C. M.P. Studio 
negative carbons. Fig. 1 indicates the marked reduction in sound- 
level that can be obtained when the C.C. M.P. Studio negative car- 
bons replace the Orotip negative carbons. 



The use of flaming arcs for furnishing general illumination in 
motion picture studios has been common practice for many years 
and their electrical and radiation characteristics have been described. 10 
The earlier carbons, which were l /% inch in diameter, were super- 
seded 11 about five years ago by copper-coated carbons 8 mm in 
diameter, the "National" M.P. Studio carbon. These carbons, 
in conjunction with the new lamps developed at that time, 4 gave 
steadier operation. Recently another radical improvement 6 has 
been made in lamp design as reported by Mole and it is now possible 
to control more closely the arc current and arc length. In order to 
give the maximum increase in steadiness of which this new mechanism 
was capable, it was desirable to design a new carbon trim. This com- 
bination of improved lamp design and new carbons gives a very 
steady light, improved color-characteristics for Technicolor photog- 
raphy, and a noticeable reduction in noise. This new trim for use 
in the "Duarc" M.R. Type 40 and similar lamps 12 consists of an 
8-mm upper positive carbon and a 7-mm negative lower carbon. 
Characteristics of this new trim are discussed below. 

Color. When Type 90 and Type 170 spotlamps are used in Techni- 




color photography, it is customary to ilhnninate the set with 
these lamps through a Hgfrt, straw-colored gelatin fitter* known as 
the Y-l fitter, and this combination results in a light-source of satis- 
factory color-characteristics. The new 8-mm 7-mm trim for 
designed to be used without a color-filter, in the 

new broadside lamps. This combination of new carbons and lamps 
gives the same spectral quality of light as the Type 90 lamp and 
Type 170 lamp plus the Y-l filter. This is best illustrated by spectral 

shown in Fig. 2, give the 

optical system. 

energy at the different visible and near- 

i "*^ liylif-lipgwiijg finiii ili^ complete 
The ordmates of Fig- 2 are in arbitrary !?, and the 
heights of the three curves have been adjusted for purposes of com- 
parison to the same value at 5440 Angstroms. These spectral-] 

optical system of the spotJamp including the Y-l filter and of the 
side arc lamp without any fitter, in order to obtain values characteris- * 
tic of the light as received on the set. The burning conditions wercl 

111 1114 TIT ill fm lln Tj|ii HI niillniii 1 ID 11141111 

and 37 arc volts for the Type 4O side arc lamp. These curves JMJM 

Oct., 1939] 



that the spectral quality of the light from the new 8-mm 7-mm 
M.P. studio carbons in the Type 40 broadside lamp closely approxi- 
mates that from the 13.6-mm H.I. carbon in the Type 90 lamp plus 
the Y-l filter. As shown in the figure, these new carbons in the 
Type 40 lamp give relatively less energy in the blue and more in the 
red than the 8-mm trim of "National" M.P. Studio carbons. 

When we speak, perhaps loosely, of color as applied to the light 
used for motion picture photography we refer, of course, to the effect 
of the light on motion picture film rather than to the effect of the 
light on the human eye. Consequently, the conventional nomen- 

8-Mm M.P. Studio Trim 

New 8-Mm 7-Mm Trim 

FIG. 3. 

Light-steadiness curves; MR-40 Duarc lamp; 40 amperes, 37 arc 

clature of color, based on average eye sensitivity, is meaningless in 
this connection, except as it may serve as a relatively crude compara- 
tive measure of light-sources of essentially similar spectral energy 
distribution. The spectral energy distribution of the light-source 
provides the only color data of significant importance in this applica- 
tion. However, to facilitate rough comparisons with other sources, 
the trichromatic coefficients and the color-temperatures have been 
calculated and are shown in Table I. 

The reduction of color- temperature from 5680K. with the "Na- 
tional" 8-mm M.P. Studio carbons to 4700K. with the new 8-mm 
7-mm trim illustrates how the color of the light from the carbon 
arc can be altered to give the quality desired. It is evident from 

380 JOY, Lozrefc, AND ZAVESKY [j. s. M. P. E. 

these results that the studios have available for color photography 
sources of illumination of the same color quality for both broad- 
side illumination and spotlighting. 


I.C.I. Trichromatic Coefficients and Color-Temperature of Light from Spotlamps 

and Broadside Lamps 

Trichromatic Color 
Coefficients Temp. 
Lamp Carbons Current Voltage x y K 

Type 90 Plus 

Y-l Filter 13.6-mm H.I. 115 57 0.349 0.357 4820 

Type 40 New 8-mm 7-mm 

N.P. M.P. Studio 40 37 0.353 0.350 4700 

Type 40 8-mm 8-mm N.P. 

M.P. Studio 40 37 0.328 0.341 5680 

Steadiness. The new 8-mm 7-mm M.P. Studio trim is de- 
signed to take full advantage of the automatic motor feed arc control 
of the recently developed lamps 6 * 12 and therefore when burned 

Ground 8-mm M.P. New 8-mm 7- Ground 

Noise Studio Trim mm Trim Noise 

FIG. 4. Effect of new 8-mm 7-mm trim in reducing noise-level in arc lamps; 
MR-40 Duarc lamp, at 40 amperes, 37 arc volts. 

in these new lamps gives a much steadier light than was heretofore 
possible. Fig. 3 shows the light steadiness curves obtained on the 
Mole-Richardson Type 40 motor-fed lamp with the older 8-mm 
upper and lower carbons and with new 8-mm 7-mm trim. This 
clearly indicates the steadiness and the superiority of this new trim 


over the older one. A comparison of these curves with the older 
type lamps and carbons as given last fall by P. Mole 6 shows the 
substantial progress over the last few years. 


The steadier burning qualities of the new 8-mm 7-mm M.P. 
Studio carbons have at the same time markedly reduced the sound- 
level, particularly the frying noise associated with the arc. Fig. 
4 shows comparative records of the sound-level of the National 
8-mm 8-mm trim and the new National 8-mm 7-mm trim 
burned in the new MR-40 lamps. These records were obtained 
by the method previously described in this paper and show clearly 
the superiority of the new trim. 

The combination of the new 8-mm 7-mm trim and the new 
lamps has eliminated the occasional disturbances which resulted in 
changes in intensity of the light, in changes in color of the light, and 
in slight audible disturbances. 


1 HANDLEY, C. W. : "Lighting for Technicolor Motion Pictures," /. Soc. Mot. 
Pict. Eng., XXV (Nov., 1935), p. 423. 

2 HANDLEY, C. W. : "The Advanced Technic of Technicolor Lighting," /. 
Soc. Mot. Pict. Eng., XXIX (Aug., 1937), p. 169. 

3 Report of the Studio Lighting Committee, /. Soc. Mot. Pict. Eng., XXVIII 
(Jan., 1937), p. 32. 

4 MOLE, P.: "New Developments in Carbon Arc Lighting," /. Soc. Mot. 
Pict. Eng., XXII (Jan., 1934), p. 51. 

6 RICHARDSON, E. C. : "Recent Developments in High-Intensity Arc Spot 
Lamps for Motion Picture Production," /. Soc. Mot. Pict. Eng., XXVIII (Feb., 
1937), p. 206. 

6 MOLE, P.: "The Evolution of Arc Broadside Lighting Equipment;" /. Soc. 
Mot. Pic. Eng., XXXII (April, 1939), p. 398. 

7 "Handbook of Colorimetry," A. C. Hardy, Editor; Color Measurement Labo- 
ratory, Mass. Inst. of Tech., Cambridge, Mass. 

8 BOWDITCH, F. T., AND NULL, M. R. : "Selected Ordinates for Computing 
Trichromatic Coefficients and Candle-Power of a Light-Source," /. Opt. Soc. Amer. 
XXVIII (Dec., 1938), p. 500. 

9 JUDD, D. B.: "Estimation of Chromaticity Differences and Nearest Color- 
Temperature on the Standard 1931 I.C.I. Colorimetric Coordinate System," 
J. Opt. Soc. Amer., XXVI (Nov., 1936), p. 421. 

10 JOY, D. B., AND DOWNES, A. C.: "Characteristics of Flame Arcs for Studio 
Lighting," Trans. Soc. Mot. Pict. Eng., XII (1928), p. 502. 

11 JOY, D. B., BOWDITCH, F. T., AND DOWNES, A. C.: "A New White-Flame 
Carbon for Photographic Light," /. Soc. Mot. Pict. Eng., XXII (Jan., 1934), p. 58. 

Internal. Phot. (Dec., 1938). 


Summary. The design requirements for this type of unit and how these require- 
ments were met in the selection of truck, body design, equipment layout, etc., are dis- 
cussed. The recording equipment utilized together with the power equipment and 
other special features of the unit are described. This type of unit has been in successful 
operation without revision. 

Before designing mobile sound-film recording units to be supplied 
to Republic Productions, Inc., engineers of that organization met 
with engineers of RCA Manufacturing Co., Inc., to formulate the 
design requirements for these units. Briefly, the following require- 
ments were established : 

(1) The mobile unit was to contain all primary power equipment and a com- j 
plete film -recording system suitable for either location or studio recording. 

(2) Maximum accessibility for service and maintenance was required. 

(3) Maximum convenience of operation was stipulated in order to make the i 
unit useful for fast production shooting. 

(4) All tires and axles were to be correctly loaded in accordance with factory 

(5) An easily maneuverable truck was required permitting fast moves in a , 
narrow studio alley, public highway, or rough location country. 

(6) The unit must be easily ventilated and well insulated against the sun's 

Tentative full-scale layouts were made on the plant floor using; 
white adhesive tape. After suggestions made by operating personnel j 
were reviewed, various alterations were made to the original layout. ; 
A full-scale model was then built using a framework of 1 X 1-inch ; 
wood strips, to which wrapping paper was nailed for walls and parti- j 
tions. After further changes suggested by operating personnel, the 
full-scale model was lifted to a truck chassis in order to check thej 
overall appearance of the unit. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
March 14, 1939. 

** RCA Manufacturing Co., Hollywood, Calif, 
t Republic Productions, Inc., North Hollywood, Calif. 



Following this preliminary work, a truck chassis was selected. 
It had early become apparent that the cab-over-engine or cab-forward 
type of truck would provide maximum body length with minimum 
wheel-base. This arrangement seemed best to fulfill the require- 
ments mentioned above. The cab-forward type of truck was finally 
selected in favor of the cab-over-engine truck since it was believed 
the former provided a less fatiguing ride for the driver and because 
less load was placed on the front axle. It was believed that this would 
increase ease of handling in rough location country. 

Our next requirements were interrelated. Ease of operation indi- 
cated head-room of approximately 6 feet 4 inches, together with a 
reasonably spacious body. Therefore, our problem was to choose a 
cab-forward truck having the greatest ratio of load-carrying-length 
to wheel-base, accompanied by an engine-mounting and drive-shaft 
construction which would allow the maximum portion of this load- 
carrying-length to have a drop-frame construction applied. These re- 
quirements indicated a passenger-bus type of chassis with a dropped- 
frame construction. The other requirements were the more con- 
ventional ones such as: adequate engine horse-power, proper tire 
size, type and size of brakes, and satisfactory general construction. 

These specifications were met within the budgeted price for the 
truck by employing a cab-forward Studebaker truck with dual rear 
wheels delivered complete with cab. Following delivery, the frame 
was dropped 9 inches throughout the portion of the chassis between 
the rear of the cab and the front spring shackles of the rear springs. 
This change did not reduce the minimum road clearance of the unit. 
Because of the construction of the truck, it was possible to run the 
drive-shaft under the floor without the usual "tunnel." Plans cover- 
ing the revised frame construction were filed with and included in 
the original warranty by the automobile manufacturer. 

Actual body design followed, and culminated in the choice of hard- 
wood and metal truck construction. The hardwood frame is screwed 
and glued together with all joints reenforced by metal strips. The 
entire body has a 2-inch fill of heat-insulating material which prac- 
| tically eliminates heat transfer from the sun. The insulation is 
sufficient for a future installation of an ice air-conditioning system. 
In order to enhance the external appearance of the truck, extra-wide 
metal sheets were used eliminating molding usually employed to 
cover the joints of two standard size sheets of body metal. In 
order to add to the appearance of the truck and avoid the usual 



U. S. M. P. E. 

difficult- to-clean space between the'cab and the body, the body was 
folded around the rear of the cab with a suitable gasket to prevent the 
entrance of dust. 

The completed truck is shown in Fig. 1. The side door shown in 
the picture is one of the two doors giving easy access to the recording 
room. Fig. 2 is a floor plan of the truck interior showing the record- 
ing room, which is directly behind the cab, and the power room which 
is at the rear of the truck. Cross- ventilation is secured by windows 
in the entrance doors, and exhaust ventilation is provided by a 

FIG. 1. The mobile sound-film recording unit. 

ventilating fan built in the roof of this compartment. The interior 
walls and ceiling are surfaced with an asbestos board with a glazed * 
finish of pleasing design. The floor and equipment decks are covered ; 
with linoleum. Nickel hardware and brushed nickel trim are used 

Space against the front wall of the recording compartment is 
utilized for film, magazine storage cabinets, and the motor-generator 
batteries. This portion of the truck interior is visible in Fig. 3. 
Fig. 3 shows the large amount of head-room above the storage bat- 
teries for servicing, together with the two lights provided. Not seen 
in the picture are two louvres for ventilation of the battery compart- 
ment. To provide adequate drainage of the battery compartment, 

Oct., 1939] 



the floor is sloped to a lead-pipe drain which empties below the truck 
; mechanism. The batteries are set on an elevated wood grid which 
I allows any spilled liquids to reach the drain. The interior of this 
i compartment and the wood grid are heavily sprayed with a coat of 
acid-resisting asphaltum. The removable door of the compartment 
| is clamped in an air-tight position, preventing acid fumes from enter- 
! ing the recording compartment. 

Above the battery compartment and shown in Fig. 3 is a light- 
tight film magazine loading compartment. Access for loading 
through the dual door construction is by means of the shielded arm- 
holes. Easily accessible and above the loading compartment is 
provided a film-storage cabinet for 30,000 feet of film. 




























FOR ( 


FIG. 2. Truck floor plan. 

Adjacent to these compartments is a closed partition-separated 
pace for storage of film magazines. To the right of this space, 
hown in Fig. 4, is a compartment which is shelved and divided into 
eparate spaces for microphones, preamplifiers, mixer panel, and 
pare tubes. All the storage compartments are felt-lined to prevent 
amage to the equipment. 

Fig. 5 shows the operating section of the recording room. The 
>ower rack on the right and audio rack on the left are built on hinges 
o provide accessibility to the wiring. An equipment storage com- 
artment is located below the power rack. The small drawer ob- 
ervable below the power rack when pulled out has a cover which 
cts as a writing desk. The monitor speaker is visible above the 
ecorder magazine. The vertical portion of the wall back of the 



.s a 



a. a 




recorder is hinged, allowing access to the rear of the recorder from the 
power compartment. Illumination for this section of the truck is 
supplied by Lumaline lamps behind opal glass mounted flush in the 

FIG. 5. Operating section of the recording room. 

All operating controls have been located so that the recordist can 
lin seated at the recorder and have all necessary audio controls 
it his left hand and all power controls at his right hand. This lay- 
nit permits ease of operation and speeding up of production. Be- 

mse of the accessibility of all controls the recordist can easily turn 



off the motor-generator between shots when on location, thereby re- 
ducing the drain on the storage batteries. 

A view of the power room from the rear of the truck is shown in 
Fig. 6. The central door is provided with a large movable window 
allowing the power compartment to be used by the mixer for "truck- 
ing shots" or under poor weather conditions. Both side doors pro- 
vide access to the cable compartments and are noteworthy since the 
door catches are operated from the inside of the truck only, preserving 
the smooth contour of the truck when these doors are closed. The 

FIG. 6. View of the rear of the truck. 

smoothness of the truck exterior was helped further by shaping these 
doors in such fashion that their hinges were combined with the normal 
"break" between the sides and rear of the truck which is usually 
covered by oval molding. The cable reels shown in the left-hand 
compartment can be cranked from the ulterior of the truck with a 
double-ended crank which makes it possible to select any one of the 
four reels for individual cranking. Directly below the cables is lo- 
cated a plug panel for the connection of all external cables. The 
cables are fed through a sponge-rubber-lined chute which allows 
closing the cable compartment door even when cables are connected 
to the plug panel. 

Oct., 1939] 



Figs. 7 and 8 show the location of equipment in the interior of the 
power room. The floor of this room and the deck over the recorder 
and amplifier-filament storage-batteries are covered with linoleum. 
This deck is used as a seat for the mixer when he works in the interior 
of the truck. Since the deck is of the same height as the motor- 
generator compartment alongside it, easy removal of the motor- 
generator for servicing is thus facilitated. Other equipment visible 
in these two pictures are the dynamotor and filter which supplies 



FIG. 7. FIG. 8. 

Views of the power room equipment. 

all B voltage, the high-capacity tungar charger for charging all stor- 
age-batteries, the motor-generator for supplying camera and recorder 
power, and the dynamotor for moviola playback supply, together 
with the various motor-starting relays. 

The truck is equipped to work with conventional microphones, and 
utilizes a compact portable mixer mounted on a simple collapsible 
stand with casters. External and internal views of the mixer are 
shown in Figs. 9 and 10. A split hinge cover not shown in the photo- 
graphs can be hooked on the side of the mixer and used for a writing 
shelf. The mixer is provided with an interphone, high-quality 



tf. S. M. p. E. 

monitor phones, adjustable dialog 'equalization, and an indirectly 
illuminated meter type volume-indicator. A talk-back microphone 
built into the mixer for rapid communication with the recordist can 
also be used for recording announcements directly on the sound- 

FIG. 9. ( Upper) External view of the mixer. 
FIG. 10. (Lower) Internal view. 

track. Preamplifiers and all connections to the truck are plugged 
into the mixer panel by appropriate cables. 

Fig. 1 1 shows the transmission diagram of the audio equipment in 
the truck. This is a bridge bus circuit with one bridging amplifier 
driving the recorder and a second bridging amplifier supplying vol- 
ume-indicator, monitor speaker, and monitor headphones. Exponen- 

Oct., 1939 J 



tial noise-reduction is provided, and an electronic volume compressor 
is a standard item in the truck. All jack circuits are normalled and 
the jack fields have been carefully arranged to resemble the trans- 
mission diagram in their physical layout. At the same time, separa- 
tion of different audio levels has been achieved. 

The recorder, powered by a three-phase synchronous motor, is 
a studio type RCA unit with an electromagnetic film drive. It is 
arranged for either standard or push-pull duplex ultraviolet variable- 
area recording. A temperature-compensated exposure-meter is 
built into the recorder. A built-in negative exposure unit is provided 
for track exposure at the beginning of each take. A photographic 
slater records scene and take numbers in the sound-track area and is 
operated in conjunction with a mechanical punch, providing the 

FIG. 11. Transmission diagram. 

identification necessary for the pre-selection of negatives. A self- 
engaging magazine take-up is provided. 

The power source for location work consists of a 120-volt bank of 
105-ampere-hour storage-batteries. These batteries drive a 0.75- 
kw motor-generator set delivering 220-volt, 3-phase, 60-cycle power 
for the recorder, cameras, and playback motors; a B supply dyn- 
amotor generating 250 volts, 300 milliamperes d-c; and a 110-volt, 
60-cycle, 225-watt single-phase a-c dynamotor for playback amplifier 
and lamp supply. When external sources of a-c are available, the 
motor-generator set and J. 10- volt, a-c dynamotor are not used. 

At times it may be necessary to supply battery power for long 
shooting schedules which do not allow sufficient time for proper 
charging of the batteries for the next day's "shooting." To provide 
for this emergency provisions have been made to "float" the power 
batteries on any suitable d-c source available, such as that normally 



[J. S. M. p. E. 

used for "booster" lights. Provisions have also been made in the 
system when external a-c is available for floating the tungar charger 
across the 8- volt, 210-ampere-hour, amplifier-filament batteries and 
the 16- volt, 210-ampere-hour, exposure-lamp batteries while recording. 
The charger utilizes six 6-ampere tungar bulbs. As is usual with 
this type of device, the nominal 36-ampere rating can be exceeded for 
a short period of time in accelerating the charging of exhausted bat- 
teries. An interesting design feature of this charger is the specially 
wound power transformers which allow the charger to be used either 
on 110- volt, 60-cycle, single-phase; or, by simply throwing a switch, 
to be fed from a 220-volt, 3-phase, 60-cycle power supply. This is 
accomplished by utilizing three separate power transformers with two 

M-l C-8 U-7 

C-2 L-3 C-3 L- 




h o* OOQ ve 

H-O/ pQQ IP i 

io* OOP '-o 9 o' OOP '-09 

^"* * " 

T T 

~ NO. 1 OUT 


L-5 R-l C-5 


~ NO 2 OUT 
_O 04ZA9I60V 

_O 250V 



-o 250V 

L-6 C-6 

_O 054 A e>240V 

FIG. 12. Filter for B supply dynamotor. 

primary windings on each. For 110- volt charging, all six primary 
coils are paralleled. For 220-volt, 3-phase charging, the two primary ; 
coils of each transformer are connected in series and the three sets of 
these primary coils are then connected in delta across the 3-phase 

The d-c starting mechanism for the motor side of the motor-gener- 
ator set is a relay-type four-point starting-box. 

Fig. 12 shows the B supply filter. This filter is of interest because 
it enables the single dynamotor unit to supply B voltage to all parts 
of the recording system. By providing filtering appropriate to the 
signal level in the circuit, this filter provides excellent voltage regula- 
tion with a negligible amount of cross-modulation between units. 
This is accomplished in spite of the fact that the current supplied to 


the noise-reduction unit is constantly fluctuating during recording 
over a range of approximately 25 milliamperes. 

Mobile film-recording systems of this type have been in use by 
Republic Productions, Inc., for several months and have amply 
justified the care spent in designing them. The authors wish to 
acknowledge their indebtedness to Mr. James L. Fields, of the RCA 
Manufacturing Co., Inc., who was in charge of construction of these 
units. We also wish to take this occasion to thank Mr. Daniel J. 
Bloomberg of Republic Productions, Inc., and the many other 
members of the RCA and Republic organizations who contributed 
valuable suggestions to the design of this mobile system. 


C. R. DAILY** 

Summary. The use of an a-c polarizing potential on a gas type PEC produces 
an alternating output current as compared with the continuous current obtained with 
a d-c polarizing potential. If the output voltage from the cell is suitably connected to 
a conventional audio-frequency amplifier and copper-oxide rectifier meter, the equip- 
ment may be conveniently used as an exposure meter. When used for the line-up of 
variable-density light-valves, the proper bias current, lamp adjustment, and lamp 
current may be readily determined. Variations of valve spacing and lamp current 
are also indicated. Numerous other applications for such a meter are being con- 

The determination of absolute or differential exposure must be 
made frequently in connection with recording sound on film. An 
experimental exposure-meter of a type not commonly employed will 
be described, which may be applied to the PEC monitor system of a 
variable-density light- valve recording channel. The device is useful 
for the determination of the required noise-reduction bias current, 
lamp current, changes in valve spacing, and other quantities which 
can be detected by changes in light-intensity on a photoelectric cell. 

At the present time adjustments of lamp position and current are 
checked by a number of methods including exposure meters, either 
with or without direct-current amplifiers, gain-frequency tests 
through the valve and the PEC amplifier, as well as routine film ex- 
posures. The methods used for the determination and adjustment 
of noise-reduction bias currents include (a) light-interrupting methods 
such as tone wheels (light choppers) ; (b) proportional bias based on 
closure current, which is usually determined by visual means; (c) 
stroboscopic methods; (d) harmonic observations, either aurally or 
by meter; (e) exposure-meters, direct reading on a PEC or in connec- 
tion with a direct-current amplifier; (/) calculated values based on 
tuning and spacing. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif.; received April 
14, 1939. 

** Paramount Pictures, Inc., Hollywood, Calif. 




Light-interrupting methods in connection with photoelectric cells 
provide a carrier which is changed in amplitude with changes in the 
amount of light falling on the cell. The use of an alternating current 
for the polarized potential of the photoelectric cell likewise provides 
a carrier. 1 If precautions are taken to reduce the capacity of the 
PEC system, a reasonable degree of linearity of a-c output from the 
cell may be obtained as a function gf light-changes applied to the cell. 
In this paper the application of an a-c polarized gas type PEC will be 
described. The cell, acting as a d-c to a-c converter, may be con- 
nected to a conventional audio-frequency amplifier and full-wave 
copper-oxide rectifier meter and used to measure changes in light 
falling on the PEC. 



FIG. 1. Schematic diagram of an exposure meter consisting of an a-c 
polarized gas type PEC, audio-frequency amplifier, and rectifier type 

Fig. 1 is the schematic diagram of a PEC amplifier which has been 
modified for use as an exposure meter. The low-voltage, low-fre- 
quency, a-c polarized potential is applied in series between the 
cathode and ground. The shape and magnitude of the alternating 
PEC current generated depend upon the voltage applied, the type 
of cell used, the load resistance, the shielding of the cell, and the 
amount of light falling on the cell. Satisfactory operation has been 
obtained with 15 to 30 volts applied to the cell and with a load resis- 
tance of 0.5 megohm. 

When this arrangement is used as an exposure meter for the deter- 
mination of bias currents, the gain of the system is first adjusted 



[J. S. M. P. E. 

until a convenient reading is obtained on the ouput meter while the 
valve is unbiased and unmodulated. Applying bias to the valve 
then reduces its spacing, the light falling on the PEC, and the mea- 
sured output current. 

The results of a series of such tests are shown in Fig. 2. The re- 
duction in measured output, in db, is plotted as a function of the de- 
sired bias current expressed in the same units. The proper bias cur- 
rent was determined from closure tests which agreed satisfactorily 





15 20 



FIG. 2. Experimental calibration of exposure meter. 
Measured reduction in output with bias current applied 
to a light-valve vs. the desired noise-reduction. Data for 
three PEC's of the same type. 

with film data over a range of 12 db. The solid curve represents a 
linear reduction in output, while the three broken curves show the 
reductions in output obtained with three different RCA-921 type 
PEC's. Substantial linearity is indicated over a range of 12 db, one 
cell being linear over a range of 20 db. Any non-linearity which may 
exist in a given system can be taken into account by. calibrating the 
output meter directly in terms of film measurements. 

The approximate efficiency of the system was determined by first 
measuring the output with a-c polarization of the PEC, using an un- 
biased, unmodulated valve; and second, with a normal d-c polarized 


PEC, the unbiased valve being modulated 100 per cent at a low fre- 
quency. With 30 volts of a-c polarization the measured output was 
approximately 15 db less than that obtained with valve modulation. 

This type of exposure meter may be used as an absolute indicator 
if suitable precautions are taken to stabilize the system. Tempera- 
ture and voltage variations to the PEC would have to be minimized 
and the audio-frequency amplifier should preferably be of the feed- 
back type to reduce gain variations to a minimum. With a stable 
system, direct determinations could then be made of the required lamp 
current, variations in current, variations in valve spacing, etc . Further 
work will have to be done to determine the degree of stability that can 
be obtained with this system. 

The author wishes to acknowledge the assistance rendered by 
Mr. L. W. Russell of Paramount Pictures, Inc., in originating and 
executing this project. 


1 ARTZT, M.: "Facsimile Transmission and Reception," Radio Facsimile, 
Radio Institutes Tech. Press, I (Oct., 1938), p. 159. 



Summary. A direct-reading densitometric method is described for the determina- 
tion of the approximate cross-modulation cancellation of bilateral variable-area 
prints. This method makes use of the relationship between film rectification and the 
accompanying change in the mean transmission of a modulated print. The static 
measurement tells the direction as -well as the approximate amount of the print 
density deviation from optimum. Complete measurements can be obtained with only 
a few inches of film, compared with several feet required by the routine dynamic mea- 
surement of cancellation. The method may simplify the checking of release prints. 

Several years ago it was determined that the quality of variable- 
area prints was a function of print density, condition of the negative, 
developer, printer, and other factors. 1 2 Subsequently the modulated 
high-frequency test was developed, which provided an index to quality 
in terms of a readily measured rectification component. 3 The ob- 
servation had also been made, however, that on a modulated print 
the average transmission decreased with increasing print density due 
to the increased fill-in of the wave. This change in transmission may 
be directly determined by densitometric measurements, and it is the 
purpose of this paper to present some data which indicate the order 
of magnitude of the change in terms of cross-modulation cancellation. 
Possible commercial uses for this method of print quality determina- 
tion also are mentioned. 

Fig. 1 represents a print of unbiased bilateral, 76-mil, variable- 
area track. Two types of track are shown: (a) unmodulated, and 
(b) 7000-cycle, modulated approximately 80 per cent by 400 cycles, 
the so-called cross-modulation track. The quality of such a print is 
normally determined by reproducing the modulated wave through a 
calibrated system and measuring the amplitude of the 400-cycle com- 
ponent of the wave which is caused by film rectification. Thirty 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
14 1939. 

** Paramount Pictures, Inc., Hollywood, Calif. 



db or greater suppression of this component is readily attained, a 
value which is generally indicative of satisfactory processing. Other 
factors contributing to quality such as printer contact must, of course, 
be taken into account. 

In order to determine whether direct densitometric measurements 
would be of any value in checking variable-area prints, use was made 
of a barrier type PEC densitometer which had a 
light aperture of approximately 85 X 240 mils. A 
number of prints were made from a normal negative 
which had three types of unbiased track: (a) un- 
modulated, (b) 7000-cycle, and (c) 400-cycle modu- 
lated with 7000 cycles. The cross-modulation can- 
cellation was measured in the routine manner, an 
optimum print density being indicated at a print 
density of 1.41. The mean track densities of the 
same print sections were then measured on the PEC 
densitometer, the track being centrally located with 
respect to the slit as indicated by the dashed lines 
(Fig. 1). In Fig. 2, the measured mean track den- 
sities for each of the three types of track are plotted 
as a function of print density. The measured cross- 
modulation cancellation is also plotted in the same 
figure, using the ordinate scale on the right. It 
will be noted that the same mean track density was 
obtained for the unmodulated and cross-modulated 
tracks at approximately the same print density as 
indicated for maximum cancellation, whereas the 
same mean density for the 7000-cycle and unmodu- 
lated tracks was obtained at a lower print density, 
in this case at a density of 1.06. 

In Fig. 3, the measured differential track den- 
sities of the cross-modulated and 7000-cycle modu- 
lated tracks, with respect to the unmodulated track are plotted 
against the corresponding cross-modulation cancellation. These 
data indicate that with the measuring set-up used, if the differen- 
tial density of the cross-modulated track, referred to the unmodu- 
lated track, does not exceed =*=0.02, the cancellation should exceed 
30 db. A number of other observations have indicated that the shift 
of the zero-differential density may be as much as 0.006 from 
that obtained by cross-modulation measurements. 

FIG. 1. Bi- 
lateral variable- 
area track. Un- 
modulated and 
400-cycle modu- 
lated 7000 track. 
Dashed lines in- 
dicate approxi- 
mate area meas- 
ured by a PEC 
densitometer to 
determine the 
amount of film 


C. R. DAILY AND I. M. CHAMBERS [J. s. M. p. E. 

The reasons have not been fully ascertained as to why the 7000- 
cycle modulated track apparently requires a lower optimum print 
density than that indicated for the cross-modulation track. Factors 
such as the possible non-linearity of the modulator will require inves- 
tigation, since the cross-modulation test does not fully analyze the 
performance of this type of modulating system. In general it may 
be said, however, that the dynamic method of checking processing is a 
reliable index of quality as indicated by actual listening tests on pro- 


lufeUi'*. !fr 
IP O (f O O 







10 Q 








^- - 





oo * u 
O o O 










.9 1.0 I.I 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2. 

FIG. 2. Measured densities of a section of variable-area track, 
at a number of print densities, for (a) unmodulated, (b) 7000-cycle 
and (c} 400-cycle modulated 7000 cycles. Also dynamic measure- 
ments of cross-modulation cancellation for the same prints. 

duction recordings. Therefore the densitometric method of analysis 
described here should also indicate the same result. 

No investigation has yet been made of the accuracy of (a) the 
method as a function of amplitude and percentage modulation of the 
cross-modulation track, (b) types of film stock, (c) track width, (d) 
slit dimensions, (e) condition of the printer, or (/) fogging condition of 
the developer. Limited tests on push-pull variable-area prints 
have not yielded promising results and no data are available on 

This processing check may be useful to laboratories releasing con- 
siderable quantities of variable-area track. If each roll of negative 
contained a few inches of suitable track, a rapid check could be made 

Oct., 1939] 



as frequently as found necessary to check the approximate quality of 
the print. Occasional routine dynamic measurements of cancellation 
would still be made as a check. One distinct advantage of the densi- 
tometric method is that it shows directly whether the print is light 
or dark, because the sign of the differential density changes in passing 
through optimum, a directive indication that is not given by a single 
routine cross-modulation measurement. 

The data reported here were obtained on a single-aperture densi- 
tometer. A more rapid method might consist of the use of a special 
densitometer wherein one light-beam would be split into two sections 








< "* 30 


































1 1 

.04 .03 .02 .01 .01 

MOD. < UNMOD. * 

.02 .03 04 05 

FIG. 3. Difference in measured densities of the cross- 
modulated and 7000-cycle tracks, referred to the unbiased 
track, vs. the dynamic measurement of cross-modulation 

and directed through two identical apertures to two PEC's. With 
the cells connected in a bridge circuit, balance could be established 
for the condition of equal light transmission over the two paths, this 
balance being obtained with unbiased, unmodulated track over both 
apertures. The film would then be moved until the unmodulated 
track was over one aperture and the cross-modulated track over the 
other. If the transmissions were no longer equal, the bridge would 
become unbalanced, indicating directly on the balance meter the 
direction of print density variation from optimum as well as the ap- 
proximate degree of cancellation. 



1 MEES, C. E. K.: "Some Photographic Aspects of Sound Recording," /. 
Soc. Mot. Pict. Eng., XXIV (May, 1935), p. 322. 

2 DIMMICK, G. L. : "High-Frequency Response from Variable- Width Records 
as Affected by Exposure and Development," /. Soc. Mot. Pict. Eng., XVII (Nov., 
1931), p. 766, 

3 BAKER, J. O., AND ROBINSON, D. H.: "Modulated High-Frequency Re- 
cording as a Means of Determining Conditions for Optimal Processing," /. Soc. 
Mot. Pict. Eng., XXX (Jan., 1938), p. 3. 


MR. SOLOW: In checking processing by this method, does it matter if the length 
of track scanned by the densitometer aperture is in one case equal to an integral 
number of waves and in another to an integral number plus a fraction? 

DR. DAILY: As previously mentioned, the aperture was 240 mils long, a value 
conveniently available on the commercial densitometer used for this study. With 
that length, the track could be moved longitudinally with only slight variations in 
indicated density. A longer scanning length might have been preferable to aver- 
age out variations. 

MR. KREUZER: Would there not be some danger in determining the optimum 
print value for a printer when using only a short section of film ? A particular sec- 
tion of sound-track might give very good performance but there doubtless would 
be periodic fluctuations in the machine. I should be concerned about taking one 
reading and then determining the correct density. 

DR. DAILY: Irregularities in the film print do occur, traceable to irregularities 
on the original negative, stock, and printer contact troubles. For these reasons 
this method is not entirely satisfactory since it indicates the condition of the print 
for only a limited section. The present commercial types of checks on printers 
for contact would still be indicated. 

MR. AALBERG: Which of the two methods do you prefer; or do you have any 

DR. DAILY: The dynamic method is normally used. The static method was 
developed to facilitate the determination of the direction of required changes in 
print density. 

The principles upon which this method are based have been observed before, 
but no reference appears in the literature as to the order of magnitude involved. 
The data so far obtained indicate that the static method is practicable and further 
study by organizations which handle considerable quantities of variable-area film 
may indicate that it has some commercial value in reducing the amount of film 
and time required to check the condition of prints. 


D. R. WHITE** 

Summary. A direct-reading photoelectric densitometer has been built which 
shows the density of the area being measured at a reading window. A density range 
from to 3.0 is covered with a reproducibility of approximately 0.005. A motor- 
driven circular neutral wedge is used as the balancing means and the density scale 
marked on the wedge is read by a stroboscopic flashing light. 

Many different types of physical photometers have been built 
and placed in use as densitometers in photographic work. No one 
type of instrument has yet gained wide acceptance as standard, and 
different laboratories have built different types to meet their needs. 
Photometers are usually required to measure both rapidly and 
accurately. The photometer which has been built and placed in 
routine use and is described here meets the requirements of speed and 
accuracy in a different way from the way they have been met before. 
The instrument is direct-reading, and its limit of speed essentially the 
speed limit of reading numbers on a scale. Accuracy has not suf- 
fered from this high speed, since readings are more reproducible than 
with the usual type of visual polarization instruments. 

Optical System. The optical system of the instrument shown in 
Fig. 1 was designed to fulfill reasonably closely the conditions requi- 
site to the measurement of the diffuse density of the deposit. An 
image of the ribbon filament of an exciter lamp was formed at the 
plane of the density being measured by two achromatic condenser 
lenses. This light-beam passes through a circular neutral wedge just 
before it reaches the density being measured and is collected by a 
photocell placed close behind the film, so that the sensitive surface of 
the photocell subtends a fairly large solid angle as viewed from the 
film. Since a production instrument was desired rather than a pri- 
mary standard, any failure to achieve complete integration of the 
transmitted beam would lead only to second-order errors, depending 

*Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
6, 1939. 

**Dupont Film Manufacturing Corp., Parlin, N. J. 




[J. S. M. P. E. 

upon differences in scattering power of various deposits as the first- 
order effect would automatically be eliminated by the initial calibra- 
tion. No trouble of commercial importance was found as it has been 
possible to measure satisfactorily both motion picture positive and 
negative films with one calibration. 

The rotation of the circular neutral wedge produces a periodic 
variation in the light reaching the photocell, the period of which de- 
pends upon the mechanical speed of the circular wedge and the ampli- 
tude of which depends upon the initial filament of the exciter lamp, 


FIG. 1. Optical system of the densitometer. 

the aperture of the condenser lenses, and the density being measured. 
Thus, by providing a means of flashing a stroboscopic lamp each time 
the illumination reaches some arbitrary fiducial value within a certain 
attainable range, the position of the neutral wedge at those instants 
can be determined by direct observation of a scale attached to the 

For our purposes, it was desired to separate the density due to 
exposure from the density due to absorption and reflection by the 
base and due to fog, on the basis of a simple subtraction. To make 
this simple, a mechanical diaphragm was introduced between the 
condenser lenses. With no film in the instrument, adjustments are 
made to show on the scale with the diaphragm partly closed. An 

Oct., 1939] 



unexposed area of the film is inserted, and the total reading noted. 
The diaphragm is then opened sufficiently to bring the reading to 
again and the densities due to exposure then appear directly, as 
the base and fog density has been compensated by the increase in 
optical aperture. 


FIG. 2. Circuit ot the densitometer. 



10 meg 
1 meg 

100,000 ohms 

100,000 ohms 

5,000 ohms 

100,000 ohms 

100,000 ohms 

200,000 ohms 

1,000 ohms 

1 meg 

10,000 ohms 

5,000 ohms 

100,000 ohms 

6 ohms 

T Audio transformer 

Ci 8 mf d 

C 2 0.01 mfd 

C 3 1.0 mfd 

C 4 0.03 mfd 

BI 6-volt storage-battery 

B z 180 volts; 5-batteries 

B 3 45-volt B-battery 

# 4 iVa-volt dry cell 

Bs, 135 volts; B-batteries 

B 6 22 Va volts; dry cells 

D +10Va volts 

E +13Va volts 

F +18 volts 

P 300-volt power-pack 

One special feature was introduced in the design of this diaphragm. 
Four leaves, moving radially, driven by a spiral thread, leave a 
-f -shaped aperture as shown in Fig. 1. The total length of each arm 
remains constant, but the breadth of each arm changes with the set- 
ting. This form leaves the cone of light striking the measuring 



[J. S. M. p. E. 

elements with approximately the same angular distribution. This 
refinement was introduced to avoid a change in the angular distri- 
bution of radiation such as would occur with an ordinary iris dia- 
phragm. For density measurements of non-scattering media, such 
a refinement would have no value. 

Electrical Circuit. The heart of the densitometer is the amplifier, 
the circuit for which is given in Fig. 2, and the means provided for 
the flashing of the stroboscopic tube, a General Radio Strobotron, 
which illuminates the scale, permitting an accurate reading thereof 
while in continuous rotation. 





J I 

r T 

I L 


FIG. 3. Wave shapes. 

A two-stage amplifier, connected as a d-c amplifier, i. e., with re- 
sistances for interstage coupling without condensers or inductances, 
is used to connect operatively the photocell with the tubes operating 
the Strobotron. This amplifier is designed to take the voltage pulse 
from the photocell and distort it completely, to the extent that at the 
output of the amplifier, it has become approximately a square wave 
(Fig. 3) having the special characteristic that the times of current or 
voltage change are directly controlled by the occurrence of the stand- 
ard or fiducial illumination of the photocell. One of these points is 
essentially fixed in the time of rotation of the circular wedge and is the 
time of abrupt change from maximum to minimum transmission of 
the wedge. The second time of current or voltage change corre- 
sponds to the time in the cycle when the light transmitted by the 
diaphragm, the neutral wedge and the density measured all add to a 


fixed value, permitting the standard illumination of the cell. Thus 
an oscillograph fed by the output of the amplifier shows a square wave, 
the two portions of which vary with the density measured. 

There are certain features of the photocell and amplifier circuit 
the importance of which was not at first apparent. An attempt was 
first made to use conditions of photocell voltage and coupling resis- 
tance such that the photocell current could be essentially a measure 
of the incident illumination. This would probably have been satis- 
factory for an instrument of small range, but for a density range of 
to 3.0 which was desired, it soon became apparent that the large 
current pulses, 1000 or more times the value for balance, introduced 
serious charges on the small capacities inherent in the system, with 
consequent erratic and spurious effects. These disappeared when a 
low voltage was used on the photocell, which was connected in such 
manner that high illumination tended to make the first grid positive, 
thus introducing through grid current, an effective shunt of the coup- 
ling resistor, and reducing the actual voltage changes of that grid to 
a swing between an amount sufficient to block the tube with no light 
on the photocell and a value great enough to produce saturation dur- 
ing the time of maximum illumination. Using low plate and screen 
grid voltages, high grid and plate coupling resistors can be used. 
The wave-form from this stage is not square, as one side has consider- 
able slope, but the second stage completes the squaring process ade- 
quately and furnishes sufficient power to operate the other tubes. 

Preliminary consideration had suggested that since actually a 
repetitive varying current was involved, a capacity-coupled amplifier 
could be used with greater circuit simplification. This view is in- 
correct, as the requirements for the production of a square wave 
from this type of original pulse are not met by such a circuit. 

The output of the square-wave amplifier feeds two tubes, one of 
which charges a condenser, Ca, during one part of the cycle which is 
subsequently discharged through the strobotron when that is trig- 
gered off by a pulse from the second or firing tube. The condenser- 
charging circuit is simple, consisting merely of a tube blocked during 
one part of the cycle and conductive during the other. A tube with 
rather high plate current is required since for some of the measure- 
ments only a small portion of the cycle can be devoted to that func- 
tion, and in that short time, actually only about 1 / 2 oo of a second, 
sufficient charge must be stored in the condenser to produce a normal 
flash of the strobotron. When this condition is not met, the strobo- 



LT. S. M. P. E. 

Iron may flash only alternate cycles, or even less regularly when more 
than one charging period is required for the condenser. 

The firing of the strobotron is accomplished by a voltage pulse 
(Fig. 3), applied between the cathode and one of the grids of the 
strobotron. A bias battery and a parallel resistance and capacitance 

FIG. 4. General view of the instrument. 

are used in this circuit to limit the firing to the one pulse needed, as 
the firing tube actually supplies two pulses of opposite polarity, from 
the square-wave input. These pulses are produced in the secondary 
of a transformer, the primary of which is fed by a square wave through 
a vacuum tube and simple net work. Some trouble was experienced 
with oscillations from the pulse nature of the excitation, but these 
were removed by trial -and-error use of loading resistances in the 


circuit. The resistance, RL, shown between the power-supply and the 
condenser, was also the result of trial-and-error elimination of irregular 
operation. Effectively, it prevents the formation of a continuous 
glow discharge through the strobotron. Such trouble occurred un- 
predictably and irregularly before the insertion of this resistance. 

D-c operation of the exciter lamp and amplifier filaments was found 
necessary, as a-c at these points introduced too much disturbance of a 
frequency to which the system was sensitive. The exciter lamp is 
operated from a battery floating on a rectifier charging system, but 
for the first stages of amplification, this system still left too great 
residual disturbance, so a two-battery system was installed, such that 
one battery charges while the second is in use. 

The timing precision of the strobotron flashes has been interesting 
to many seeing the machine in operation. With a rotational speed 
of 20 rps, the repetition of the flashing of the strobotron is within 
1 /i6,ooo of a second under actual operating conditions, and the 
duration of the flash can not be greater than of a second, as 
judged by the distinctness of the lines, and probably is actually ap- 
preciably less than this. 

The instrument was calibrated by the use of a temporary arbitrary 
scale on the circular wedge. Strips with known densities, as deter- 
mined by polarization visual instruments, were put in the new instru- 
ment and the corresponding scale readings noted. From these re- 
sults a final density scale was constructed and mounted. Further 
checks showed that procedure had been carried through successfully, 
and the calibration has remained constant for months. 

Mechanical Features. The mechanical design (Fig. 4) was adopted 
to facilitate the handling of test strips. The strips to be read are 
held in a revolving drum by simple clips. The shaft is tripped be- 
tween positions for successive densities by a ratchet escapement 
mechanism which is provided for either foot or hand operation. The 
diaphragm control knob is brought to the front of the machine. 

No other control requires frequent attention. Bias adjustments 
and battery checks are occasionally required, but experience has 
shown that none of these affects the reading in such manner as to 
result in minor errors, easily ignored. In practice it either reads 
correctly or not at all. Actually, servicing this densitometer has 
taken less time than maintenance of a group of visual instruments of 
similar total capacity, and the reproducibility of the readings has been 
considerably greater. 



Summary. The term "acoustic condition factor" is used as a general term 
descriptive of the acoustic environs of a point in an enclosure. Relationships ex- 
pressed as ratios are given for several quantities, such as "useful" and "harmful" 
sound, direct, and generally reflected sound energy and sound intensity. Curves 
are shown representing loci for partial antinodes produced by interference between 
direct and first as well as second reflections in a rectangular room in which the sound 
source is located symmetrically. Equations are given expressing the minimal dis- 
tance between source of sound and microphone for the probable avoidance of recording 
absolute nodes. 

In recording sound for motion pictures as well as in a general 
evaluation of the acoustics of a room one often desires to know more 
than the value of the calculated or measured reverberation time in the 
enclosure. Recordings of sound, like various acoustic measure- 
ments, are specific ; that is, they are related to a definite position of 
the microphone hi the room. Varying this position may change the 
character of the reproduced sound, may make it sound crisper, or 
more reverberant, or may enhance or lessen its intelligibility. The 
reverberation time, being practically the same at every point in the 
room, provides no explanation for this acoustic condition and should 
therefore be considered only as a summary or average factor. 

The term acoustic condition factor is used in the following as a 
general term descriptive of the acoustic environs of a point in the 
enclosure, although other names are sometimes used for one or the 
other of these "factors," as shall be noted. 

Following is a list of the symbols used : 

5 = total interior surface of room 
a = average absorptivity 
= SflxS, 

b = average reflectivity 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 1, 1939. 

** RCA Manufacturing Co., Hollywood, Calif. 





A = total absorption in room 

= aS 

V = volume of room 
/ = time 

T = time of reverberation 
P = power output of source 
c = velocity of sound 
Ed = direct energy density 
E R = generally reflected energy density 
Eo = steady state energy density 
I D = intensity due to direct sound 
I R = intensity due to generally reflected sound 
7 = steady-state intensity 
X = wavelength 

D = distance between source of sound and point of observation in room 
/ = frequency 
= solid cone of reception of microphone 

The sound-pressure at any point in a room is made up of two parts 
sound which comes to the point of observation directly from the 
source, and that part which is reflected from the walls of the en- 
closure. In the literature of architectural acoustics this latter part is 
sometimes divided into "initial" and "residual" sound, 1 the former be- 
ing all that part of the reflected sound which comes to the point of ob- 
servation within one-sixteenth of a second after it is emitted. The lat- 
ter part all the reflected sound which comes to an auditor after one- 
sixteenth of a second after emission is considered as not contributing 
materially to the intelligibility of speech, but being assistant mainly in 
establishing the character, timbre, or tone-color of the sound. 

As far as intelligibility of speech is concerned, sound in a room can 
be divided into two components, useful and harmful. The former, 
besides the initial sound, naturally also includes the direct sound, 
while the latter is the "residual" part mentioned above plus what 
unwanted sound, that is, noise, is existing in the room. This division 
of sound is due to Strutt, 2 who thus expanded Zwikker's equation 1 
dealing only with initial and residual sound in a room. Strutt's 
"acoustic condition factor" is given by: 






U. S. M. P. E. 

where E n represents the noise level' hi the room. Assuming the 
noise level to be negligibly small, as in a well-insulated recording 
studio, the above equation reduces to: 



For good intelligibility the value of this equation is by some investi- 
gators set equal to unity, although Strutt himself places it "some- 
what greater." Setting it equal to 2, to be on the safe side, as the 
saying is, we get for the distance beyond which no dialog recording 

FIG. 1. Acoustic condition factor for non-directional microphone. 

should be made, or for that matter beyond which no critical judging 
of intelligibility of reproduced sound should be done, as hi a review 


f 1030 


3 -e 

0.862 \ 

It is therefore well also hi theaters whose length is considerable to 
reduce the reverberation time a little beyond its optimal value to 
secure good intelligibility hi the rear sections of the house. 

Olson 3 showed that the effect of the ratio of direct to generally 
reflected sound energy at a point in a room is dependent upon the 


solid cone of reception of the particular type of microphone used. 
Letting ft be the solid cone of reception of the microphone, we have : 

4P(1 - a) 
_ E R 11 _ cA Q 4>s(l -a) 

ED 4* P * ~ A 

since the response of a directional microphone to generally reflected 
sound will be 0,/^.ir times that of a non-directional microphone. In 
the case of a non-directional microphone (12 = 47r) , we have 



This equation is shown graphically by Fig. 1. It illustrates the 
great preponderance of generally reflected sound at some distance 
from the source. 

Maxfield 4 has introduced an acoustic condition factor termed 
"liveness." In place of the generally reflected sound energy density 
(J3 R = 4P(1 a)/cA, that is, of all the sound remaining after the 
first reflection), the steady-state energy density is used. Thus 


^i cA = = 

E D P A V 

From experiments it was found that technicians preferred a lower 
range of liveness than did musicians and the public (that is, a closer 
recording distance or a less reverberant room, distance remaining 
constant); also, that the range of liveness acceptable to any one 
person was broad, and that technicians accepted as tolerable a 
much smaller range of liveness than did musicians or the public at 

Rabinovich 5 examined the case of the minimum perceptible varia- 
tion of distance for three different sources of sound singing, speech, 
and violin playing. By the use of the equation : 

ID+IR A + 4^(1 - a) 

he was able to show that the value of AQ 6 /(?6 (AQ 6 being the value 
corresponding to the increment in Q 6 for the related minimum per- 
ceptible variation in distance for the particular source of sound) 
was a constant for values of D not too close to the source. 



[J. S. M. P. E. 

The acoustic conditions at a point in space produced by the phase 
relationship of direct and reflected sound are of utmost importance in 
the recording of sound, and can at large recording distances constitute 
a frequency distortion factor far more influential in controlling the 
quality of recorded sound than the irregularities in the frequency 
response characteristic of the microphone used. While the inter- 
ference field in a room with regular contours can to some extent be 
determined theoretically, as shall be shown a little later, it would 
involve considerable labor to evaluate such a field, even for a simple 
case, when obstacles in the room have to be considered. 



'. 2, 3.) 

FIG. 2. Illustrating effect of reflective surface near the 

The problem of reducing the effect of sound-pressure peaks pro- 
duced by interference has, together with other factors, stimulated the 
use of so-called compressors or amplitude range controllers in the ; 
recording of sound. At the point P of Fig. 2 the ratio of the re- 
flected to direct sound pressure is given by : 


+ (H - IT? 

In the simplified case where 7=0, this equation comes to 

P R _ x(l - ay/> 
PD V* 1 4- H* 

Oct., 1939] 



Fig. 2A shows the variation of this ratio with frequency for the 
case indicated on the figure. It is seen that serious distortion can 
result in the recorded sound when a highly reflective surface is in 
proximity of the transmitter. The use of two microphones can do 
little to correct such a condition, since it is practically extremely im- 
probable that while partial antinodes exist at the position of one micro- 
phone there will be compensating partial nodes at the position of the 
other transmitter. 

Variation of ratio of reflected to direct sound with frequency. 

Fig. 2B shows the frequency response of a loud speaker as taken in 
the open and also curves obtained at two different positions in a room 
using the same measuring equipment. It is these large fluctuations 
in sound-pressure which can seriously distort recorded sound when 
the microphone distance becomes too great to suppress the influence 
of the reflected sound. While these curves were obtained under 
practically steady-state conditions, the situation is not changed 
much for transients, since it takes but a small fraction of a second 
in many a room for the sound-energy to build up within two or three 
db of the steady-state energy density. 



tf. S. M. p. E. 

The use of a compressor, therefore, can do much to reduce the 
effect of these extreme pressure peaks due to partial interference 
antinodes. A perhaps rare but nonetheless interesting instance 
where the compressor can be made a useful tool arises when a pressure 
microphone is located during a sustained tone, not at a pressure 
maximum but at a pressure minimum. With but little sound com- 
ing from the monitoring speaker, the mixer may likely open the volume 
control in the mixing console to secure a higher volume level. If 
now the direct sound (almost equal in strength but opposite in phase 
to most of the generally reflected sound) is suddenly stopped, the 








































J >' 













=/v > 










3 AK. 







r ft 





v Roosv 














/ V ^&G 

/77 2 

J . 









C M 



000 i 


' 'I0< 

FIG. 2B. 

Response of a loud speaker in the open and two different positions 
in a room. 

sound level will rapidly build up, since the opposing direct sound is | 
absent, and a noticeable sharp increase in recording level may result 
with a possible serious overloading of the recording mechanism in the 
absence of a compressor. 

It may be of value now to consider in some detail the interference 
field for a simplified and ideal case in a rectangular enclosure of 6i-> 
mensions large in comparison to the wavelength of sound. Consider- 
ing Fig. 2 and assuming a point source of sound we know that the 
sound from the reflecting surface will be in phase with the direct 
sound whenever the path difference between the reflected and the 
direct sound equals a wavelength or a multiple thereof of the fre-i 

Oct., 1939] 



quency under consideration; also, the reflected sound will be out of 
phase with the direct sound when this path difference amounts to 

i half a wavelength or an odd multiple thereof. 

There are of course an infinite number of points within the confines 

I of the sound field where the reflected sound will be in or out of phase 

FIG. 3. Loci of partial antinodes. 

with the direct sound. For any n, however (n = integer representing 
a multiple of the wavelength), there will be a locus along which this 
phenomenon occurs. As shown by Fig. 3, along any one of the 
curves marked n = 1, n = 2, etc., the reflected sound will be in phase 
with the direct sound. Attention should be paid to the fact that the 
loci of partial nodes and antinodes do not simultaneously refer to 
pressure and particle velocity a fact important to consider when 


FIG. 4. Loci of partial nodes. 

measurements are made with different types of microphones. We 
iieal here, of course, with partial nodes and partial antinodes, not only 
Because a reflecting surface is never 100 per cent sound-reflective, 
)ut also because the reflected sound has to travel a greater distance 
han the direct before it can combine with it. 

Fig. 4 shows the loci of partial nodes for the same geometric con- 
iguration used in Fig. 3 for the loci of partial antinodes. 



LF. S. M. p. El 

Fig. 5 shows "first-reflection loci" of partial antinodes for a fre< 
quency of 100 cycles for a plane through a rectangular room of di 
mensions indicated on the figure. If the room is as high as it is wide 
the loci are curves of revolution about the center axis of the room 
Points marked PI, P 2 , ^3, P*, PS are points where the reflected sounc 
from each one of the three reflecting surfaces meets the direct sounc 
in phase with the direct sound, and hence are points where a markec 
increase in pressure or reduction in particle velocity may be expected 

FIG. 5. Loci for partial antinodes produced by interference between 
direct sound and first reflections. 

Fig. 6 shows a few of the possible second-reflection loci of partid 
antinodes for the same room. Second reflections still constitute a: 
important factor in establishing the space interference field for steady 
state conditions, although they may be less important when 
field in a large room assumes a transient character. 

Frei 6 has shown that for a sound-source located within a reel 
tangular room, absolute nodes at a distance D from the sound-sourc 
are not probable if the mean reflectivity of the room is of the orde 
of a factor <2e given by : 

= 2V 




This means, for instance, that in the case of a sound-stage of 
volume 100,000 cubic-feet, having a total interior surface of 14,000 
square-feet, the mean absorptivity should be no less than 0.5 if we 
wish to count with the probability of not observing nodes at a record- 
ing distance of D = \/V/2 = 24.8 feet. For shorter recording dis- 
tances, obviously, the average absorptivity need not be so large. At 
10 feet from the source, this average absorptivity can be of the order 
of 0.20. 

J..S///////' /////' '//////////////' '////// '/S/// /////SS "/////////'// '/SS/r 

FIG. 6. Loci for partial antinodes produced by interference between direct 
sound and second reflections. 

It should be remembered that a partial node or antinode for 100 
ycles is also one for multiples of that frequency 100, 200 cycles, 
tc. The exact magnitude of the total pressure at points of such 
>artial nodes or antinodes will not be the same for all multiples of 

fundamental frequency, since the average absorptivity of the sur- 
ace of a room is generally a function of frequency. 

Recently interest has been revived in the work started by Wente, 7 

ho plotted the difference in the irregularities of a "sound trans- 
nission curve" against the amount of absorption in the room and 
btained a smooth function. Hunt, 8 by means of a small-model 
tiamber, was able to show qualitatively the effect on the acoustic 


transmission curve of placing sound-absorbing material in the cham- 
ber, and also determined quantitatively the variation of absorptivity 
of the acoustic material with angle of incidence of the sound. 

It is possible that such sound transmission measurements will 
some day give us a factor representing the degree of sound diffusion in 
a room. Such a factor would certainly be helpful toward a more 
comprehensive evaluation of the acoustics of an enclosure. 


1 ZWIKKER, C. : "Verstaanbarkeit von luidsprekerinstallaties," De Qugenieur. 
44 (1929), No. 39. 

2 STRUTT, M. J. O.: "Raumakustik" (Handbuch der Experimental physii 
von Wiens-Harms), 17 (1934), p. 460. 

STRUTT, M. J. O.: "On a Physiological Effect of Several Sources of Sound 
on the Ear and Its Consequences in Architectural Acoustics," /. Acoust. Soc. on 
Amer., 6 (March, 1935), p. 155. 

3 OLSON, H. F.: "The Ribbon Microphone," J. Soc. Mot. Pict. Eng., XV1J 
(June, 1931), p. 695. 

4 MAXFIELD, J. P.: "Some of the Latest Developments in Sound Recording 
and Reproduction," Tech. Bull. Acad. Mot. Pict. Arts & Sciences (April, 1935). 

MAXFIELD, J. P., COLLEDGE, A. W., and FRIEBUS, R. T. : "Pick-Up for Sounc 
Motion Pictures (Including Stereophonic)," /. Soc. Mot. Pict. Eng., XXX (June 
1938), p. 666. 

5 RABINOVICH, A. V.: "The Effect of Distance in the Broadcasting Studio,' 
/. Acoust. Soc. Amer., 7 (March, 1936), p. 199. 

6 FREI, H.: "Elektroakustische Untersuchungen in Hallraumen," Fran 
Deuticke (Leipzig), 1936, p. 34. 

7 WENTE, E. C.: "Measurement of Room Acoustics," /. Soc. Mot. Pict 
Eng., XXVI (Feb., 1936), p. 145. 

8 HUNT, F. L. : "Investigation of Room Acoustics by Steady-State Trans 
mission Measurements," J. Acoust. Soc. Amer., 10 (Jan., 1939), p. 216. 



Summary. This paper avoids technicalities and formulas, reaching back to 
elementary acoustics which are often side-tracked. Controlled reflection plays the 
leading role, with the minor parts delegated to sound diffusion and uniform energy 
distribution. Although much can be mathematically proved, the only satisfying 
conditions are the apparent ones, which are judged and gauged by the normal human 

Audio effects due to the physical characteristics of both sound absorbents and 
building materials are explained and their proper locations emphasized. Although 
a room can have the desired optimal reverberation time over the entire frequency 
response characteristic, it can still be unsuitable for the rendition of speech and 
music that is clear and distinct; the shape, size, and contours of the six surfaces in a 
room, plus the incidental equipment and purpose, are the deciding factors on how 
much and where the reflecting and absorbing materials should be placed. 

Instead of composing a technical paper containing many formulas 
and mathematical proofs, a brief discussion will be presented on some 
elementary facts of acoustics which, because of their obviousness, are 
often completely overlooked. In light of new experiences and ex- 
periments, an occasional review of fundamentals, by reduction to the 
simplest forms of analogy, will bring us closer to the correct solution 
and appreciation of our daily problems in acoustics. 

Sound is a complex form of energy. Conversion to other forms of 
energy, such as heat or mechanical energy, forms the usual method of 
control. In some respects it obeys certain basic laws of optics but, 
due to its very much longer wavelengths, it can not be as readily 
collected and focused in a direct path as can a beam of light. Sound 
bends around barriers and corners in a manner similar to eddy cur- 
rents in fluids but, if its path in a certain type of room enclosure can 
be predetermined, it is possible to prevent many of the undesirable 
conditions which result in the average auditorium or room. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
1, 1939. 

** Acoustical Engineering Co., Los Angeles, Calif. 


422 C. M. MUGLER [j. s. M. P. E. 

Proper acoustic control is accomplished by the consideration and 
satisfaction of the following: 

(1) relation of reflected and direct sound-waves; 

(2) uniform energy distribution; 
(5) diffusion; 

(4) reverberation period ; 

(5) room dimensions; 

(6) physical contouring; 

(7) types and character of reflecting and absorbing surfaces; 
(5) purpose and intended utility of the room. 

Reflected sound is the most important consideration in this group 
reflected sound which is controlled to certain definite limits. 

In an enclosure wherein the ears receive the direct waves only, the 
rendition of speech and music is lifeless and without character. 
This condition is created outdoors or by the treatment of all surfaces 
with a low absorption material, yet the reverberation characteristics 
could be within optimal time limits. In other words, uniform distri- 
bution of an absorption material will not produce acceptable acoustic 
conditions. Conversely, we can not install at random on the sur- 
faces of a room a highly absorptive material and be assured of ac- 
ceptable acoustic conditions. 

Reverberation formulas are of little value in the proper design of 
acoustic correction, except as a check. All reverberation time for- 
mulas are based upon two fundamental assumptions which have 
proved to be fallacies : 

(1) they assume uniform distribution of sound energy; 

(2) that absorption is proportional to the area of the absorbent. 

Although it is not advisable to disregard absolutely the data obtain- 
able from this time-honored method of calculation, the sole considera- 
tion of acoustic problems upon this basis would certainly lead to 
unsatisfactory acoustic treatment. Emphasis must be placed upon 
the proper selection of a sound absorbent and its proper and strategic 
location with the subject room. 

A room has acceptable acoustic conditions when the rendition of 
speech and music is characteristically reproduced and is clear and 
distinct. It is the controlled reflection which builds up the energy 
and gives character and intelligibility to this speech and music 
"brilliancy," implying that speech and music possess depth and 
properly sustained energy impulses without the "dampened" effect. 
"Brilliancy" is not produced by, or associated with, any degree of 


"liveness" or "deadness" in a room. "Brilliancy" concerns the 
maintenance of the initial energy impulses at a high level for a longer 
period of time by controlled reflection without exceeding the optimal 
reverberation time. Such a room has the apparent effect of longer 
reverberation time without interference, standing-wave patterns, 
etc., and, therefore, gives sharply defined clarity of tone to speech and 

Conventional rooms or auditoriums can be fundamentally analyzed 
as follows: each room contains three sets of parallel reflective sur- 
faces: i. e., floor and ceiling, side walls, and end walls. There is only 
one logical solution for the elimination of standing waves, flutter 
echoes, and phase distortion, and for building up and uniformly dif- 
fusing the sound energy in a room : 

(a) the side walls require the installation of a definite amount of sound- 
absorbing units calculated for all audible frequencies ; 

(b) the end wall opposite the source of sound must likewise be treated ; 

(c) as the floor is generally carpeted and contains seats, the ceiling which 
is opposite will not require any acoustical treatment in the average room. It 
should remain hard and reflective. 

This last statement is conditioned upon the time lag between the 
reflected and direct waves not exceeding one-fifteenth of a second. 
The absorbents are to be acoustical materials or systems of ma- 
terials having not less than the following sound-absorption charac- 
teristics at these frequencies which are selected at a 60-db level above 
the threshold of audibility and closely correspond to the normal ear 

response curve: 


Frequency 64 128 256 516 1024 2048 4096 8192 
Absorption 25 30 50 70 70 70 65 60 

Uniform energy distribution can be achieved only by splaying 
certain parts of the wall and ceiling areas. In order to produce 
Droper diffusion, this splaying must be carefully designed to break 
up or diverge the reflected waves. The splays must be surfaced with 
an absorbent or hard reflective material according to their locations 
n relationship to the source of sound and the size and purpose of the 

In order to prevent panel vibrations, all the surfaces, including the 
floor, walls, and ceiling, must be rigidly braced. In construction 
where panels or walls can vibrate there is always the possibility of 
induced transient distortion. Cavity resonance feed-back in the 

424 C. M. MUGLER [J. S. M. P. E. 

form of background noise is a serious condition which is found in 
many otherwise acoustically acceptable rooms. This is caused by 
the sound building up between the various surfaces of the floor, 
walls, and ceiling construction and returning at audible levels to the 
normal room interior. This is a general condition when the wall 
surfaces are furred out with lath and plaster. Even with an approxi- 
mate attenuation of 24 db, the sound transmitted into the cavity 
behind would build up in energy and be directed back as noise, and 
not as sound, into the room. 

Serious consideration must be given to the subject of sound-ab- 
sorbing materials. It is* very important to consider the physical 
characteristics of the materials and to scrutinize carefully the sound 
absorption coefficients at all frequencies, particularly in the lower end 
of the spectrum. Acoustical materials are generally divided into the 
following types : 

(1) The direct-absorbing type, where the sound energy is converted into heat 
by friction encountered in passing through the crevices and pores of the material. 
Examples are acoustical plasters and hard rigid precast tiles. 

(2) The direct-absorbing type composed of felted fibers, which absorbs sound 
in the same manner as type 1 and, in addition, gains efficiency through a dampen- 
ing effect due to the resiliency of the material. Examples are bagasse fiber tiles 
and wood fiber tiles. 

(5) The high-efficiency type, which is covered with a perforated hard surface. 
The hard surface material serves only as a mask for the unsightly absorbent be- 
hind. Example is mineral wool surfaced with perforated metal, transite, or 

Summarizing, the three types of materials are more easily de- 
scribed as (7) hard and porous; (2) resilient and porous; (3) hard and 

The effect of these three types of materials will be different even 
when they possess identical sound-absorption characteristics. Theo- 
retically and practically, the reverberation times would be 'identical 
in any three average rooms individually treated with each of these 
three types of materials, based on an equal amount of absorption 
units. However, the apparent effect upon the character and brilli- 
ance of the speech and music, as judged by the human ears, would be 
distinctly different. The effect would be directly due to the charac- 
ter of the reflected sound. 

Particular importance must be attached to the installation of these 
acoustical materials. Too often the material is installed in such a 
way that panel vibrations and the subsequent detrimental effect of 


resonating peaks are generated. The simplest case is wherein the 
material is mounted on furring strips. In most cases these resonating 
peaks can be predicted and at what frequency range they will occur, 
depending upon the type of material utilized and the method of 

Observation of the reflected sound will determine the relative 
merit of these three types of materials. Exact preference should be 
influenced by the intended use. There is little difference between 
the absorption coefficients of metal and wood but the apparent effect 
is radically different. The reflected sound from wood is clear and 
bell-like with individual characteristics of harmonics and overtones, 
whereas the reflected sound from metal is sharp and shrill. The ab- 
sorbent must be selected by studying its sound-absorption coefficients 
at each and every audible frequency. 

Although carpets and velour drapes are inexpensive to install for 
acoustical treatment, they are not suitable for critical listening be- 
cause they absorb too little energy at the pitch of the male voice and 
pitch of the female voice, and too much energy at the high frequencies. 
For instance, lined carpet absorbs only approximately 7 per cent of 
the incidental sound at 128 cps, whereas at 1024 cps it absorbs ap- 
proximately 55 per cent. Therefore, were a room treated with such 
materials, it would have a booming effect due to the masking of the 
higher frequencies by the lows. 

More attention must be given to the loudness curve and to the ear 
response characteristics. Too often the higher frequencies and their 
delicate tonal components are masked out by the energy of the lower 
frequencies. For instance, to raise a 1000-cycle note from the thresh- 
old of inaudibility to 80 db requires 80 db, whereas it requires only 
35 db to raise a 100-cycle note from the threshold of inaudibility to 
80 db. 

The remedy in preserving the naturalness of speech and music 
within a room is not in further improvements of mechanical and 
electrical equipment, but in restudying elementary acoustics and in 
choosing those materials which have selective sound absorption 
coefficients. Reverberation time formulas should be used only as a 
check against the superior methods of image projection of reflected 
sound and energy-level distribution analysis. 


Summary. When the modern optical industry was born, this country was pre- 
dominantly agricultural. Its principal industrial developments related to transporta- 
tion. It was natural, therefore, that Europe should have gained great prestige in the 
field of optics in the final quarter of the nineteenth century. 

With the turn of the century, however, agricultural developments had about reached 
their limit and industrial activity began to occupy a larger place in American life. 
Along with others the optical industry felt the incentive to greater activity and the 
first fifteen years of this century saw a rapid advance in the magnitude of the industry 
and improvement in the quality of its product. 

We are still, however, completely dependent on European sources of supply for 
our optical glass and for some of the small-demand class of laboratory instruments. 
Then came the war that not only cut off all aid from Europe but ultimately led Europe 
to our doors with appeals for optical munitions. 

The war only hastened what would have been inevitable anyway, viz., the complete 
independence of America in optical matters. 

The American optical industry has now reached a point where its raw materials 
(optical glass} and its technical skill recognize no superiors. It can make any 
practical optical element or instrument for which quantitative specifications can be 

If any justification is required for such a review as this, it lies in the 
fact that it is always a salutary procedure to pause occasionally to 
look back to see whence we have come and to look ahead to see where 
we are going. In view of the dependence of the motion picture indus- 
try upon the achievements of the optical industry such a review 
should not be out of place in a meeting of this kind where there are 
gathered together the principal technicians of the country engaged 
in both the production and projection of motion pictures as well as in 
the manufacture of equipment therefor. 

Prior to 1875 practical optics was virtually an art. It is true that 
in spite of the handicaps of an inadequate theory of image formation 
and a limited range of optical glasses men like Galileo, Leeuwenhoek, 
Dollond, Fraunhofer, Chevalier, Petzval, and Amici produced tele- 
* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 7, 1939. 

** Bausch & Lomb Optical Co., Rochester, N. Y. 



scope objectives, microscopes, and photographic lenses of astonishingly 
good quality. Each lens they made, however, had in it something of 
the nature of a painting, for example. It was in no sense a manufac- 
tured product. 

Applied optics began to develop into an industry at about the same 
time and in response to the same forces as many other activities. 
The modern optical industry did not develop because Schott learned 
how to make some new kinds of glass nor because of Abbe's contribu- 
tions to the theory of optical instruments but, on the other hand, 
they were impelled to conduct their investigations by the same condi- 
tions that produced such phenomena as Thomas Edison and Henry 
Ford. Developments in optics paralleled more startling develop- 
ments in illumination, transportation, and communications. To 
support this view it will be pointed out a little later that the same 
forces were operating at the same time in this country. 

At the time the modern optical industry was born this country was 
engaged in the gigantic task of taming a continent. Our foremost 
industry was agriculture and our second most important activity was 
to provide means of transporting our agricultural products to the 
seaboard for shipment to Europe. We had few people interested in 
the slow and painstaking task of developing the manufacture of such 
nonessentials as lenses or optical instruments. 

We did have a few pioneers, however. At a time when no one had 
even a workable theory of the microscope, Spencer and Tolles were 
making microscope objectives that were rated as equal to any pro- 
duced in Europe. Alvan Clark, Senior, was making good achromatic 
lenses and training his son Alvan Clark, Junior, who was later to 
make so distinguished a mark as the maker of some of the world's 
finest telescope objectives. The largest of these was the 40-inch 
lens of the Yerkes Observatory the largest refractor in the world and 
likely to continue to hold that honor. J. J. Bausch and Capt. Henry 
Lomb were establishing the foundations of the present Bausch & 
Lomb Optical Co. and the Wells family those of the American Optical 

At the time applied optics was beginning to develop in Europe, 
Edward Bausch in this country began to have visions of making 
microscopes in such a way that all who needed them might have them. 
Fifty years later there were more microscopes in any well equipped 
high school than there were in the whole of the United States of 
America when he made his first instrument. 

428 W. B. RAYTON [j. s. M. p. E. 

In spite of these illustrious American pioneers, the resources of 
Europe applied to the development of optics led to a much more rapid 
growth of the industry there than could be managed here and from 
1875 to approximately 1910 there is no doubt that both in theory and 
application optics in Europe led the world. The impetus supplied 
by Schott's optical glasses and by the enrichment of the theory of 
optical instruments provided by Ernst Abbe and his colleagues as well 
as the awakened demand for microscopes, optical measuring instru- 
ments, photographic lenses, and optical munitions of war resulted in a 
relatively tremendous growth of the infant optical industry during 
this period. The optical industry of America was drawn along more 
or less like the tail of a kite but remained dependent upon Europe 
for its supply of optical glass and for much of its supply of optical 

Because the American optical industry did not during this period 
attempt to meet all of the optical needs of the American laboratory 
the impression arose that American technical and scientific abilities 
were not equal to the task. This idea was nourished by the impres- 
sions gained during their post-graduate studies abroad of American 
scientific and medical students who felt that the educational institu- 
tions of this country could not offer instruction comparable with that 
obtainable in the universities of Germany and England. Using 
there for the first time optical equipment in their researches they re- 
turned to this country imbued with the idea that such apparatus 
could only be made in Europe. Perhaps some of this fixation remains 
still in some quarters but it is certainly insignificant in volume com- 
pared with thirty years ago. 

As an illustration, there was a tune some years ago when a good 
American citizen assured me when we first planned to put a spectro- 
graph on the market that Bausch & Lomb would have to make a 
better spectrograph and sell it for less money than the principal 
European maker of such apparatus before we could interest him. 
To offset his discouraging attitude we had the encouragement offered 
us by a European-born and educated chemist who told us that if we 
could make an instrument somewhere near as good as the European 
instrument and sell it for not too much more money, he would use it 
and promote its acceptance in the field of American spectroscopy. 
The outcome was an instrument that has changed the conception of 
the world as to what a spectrograph ought to be. 

While by 1910, the optical industry of America was beginning in 


some measure to parallel the achievements of Europe, it still made no 
optical glass and still failed to make many of the optical instruments 
that although tremendously important from the standpoint of the 
research laboratory were salable only in the most limited quantities. 

The outbreak of the World War changed the picture abruptly. 
The isolation and a little later the increased military needs of our 
country focused attention on three weaknesses: first, optical glass; 
second, military optical instruments; and, third, optical measuring 
instruments. The ophthalmic needs of the country were adequately 
provided for; we did not need to go to Europe for fine microscopes, 
Bausch & Lomb and the Eastman Kodak Co. with its then subsidiary, 
Folmer-Schwing Division, were able to supply the country with ex- 
cellent cameras and photographic lenses; the needs of the American 
astronomer were met with equipment second to none in the world, and 
we were making excellent surveying instruments. On the other hand, 
we depended on Europe for glass; we imported many of the optical 
elements needed in the manufacture of range finders, and we obtained 
from European sources our spectrographs, refractometers, spectro- 
photometers, colorimeters, etc. 

The story of the development of optical glass production has been 
told frequently enough to permit omitting it here beyond saying that 
we are now absolutely independent in so far as concerns the manu- 
facture of essential glasses. We no longer import optical elements 
for range finders and in so far as we have had an opportunity for 
making comparisons it is our conviction that the Inspectors of the 
U. S. Navy would not accept the product of the European manufac- 
turers and that they demand from us qualities and performance that 
are not equalled elsewhere. I regret that I am not permitted to for- 
tify this statement with more definite information. Some of the 
finest optical work accomplished lies concealed behind the wall of 
secrecy imposed with probably good reason by the military authori- 

Since the War the optical industry of this country has designed and 
now manufactures practically every kind of optical instrument re- 
quired by the research laboratory, in addition to improving and in- 
creasing production of all of the more common kinds of optical in- 

Coming a little closer to the interests of this organization, the 
production of instruments for the projection of pictures has been pre- 
eminently an American achievement. In the field of projection of 

430 W. B. RAYTON [j. s. M. p. E. 

still pictures the Bausch & Lomb Optical Co. name has become in- 
corporated in the word Balopticon recognized by the dictionaries as an 
instrument for the projection of pictures, and projectors made in 
Rochester have been sold in all parts of the world. Motion pictures 
are, of course, conspicuously an American development. From the 
beginning of the art, motion pictures have employed domestic equip- 
ment and, very definitely contradicting the statement made a few 
years ago by our friends the British Optical Instrument Manufac- 
turers Association, the great majority of motion pictures projected 
in American theaters today, yesterday, and ever since the birth of the 
art, are and have been projected through American lenses. 

For some reason not too easy to understand, the people who photo- 
graph professional motion pictures have not shared the confidence 
in American lenses manifested by those whose business it is to pro- 
ject these pictures after they are made. In view of the fact that the 
effect observed by the audience in a theater is the combined effect 
of the photography and the projection, and having further in mind 
the fact that, in some respects, the demands made on a projection 
lens today are more severe than those made on the "taking" lens, it is 
not unreasonable to suppose that the condition just stated is a tem- 
porary one and that the American cinematographer will eventually 
come to share with the microscopist, the spectroscopist, the astrono- 
mer, the ophthalmologist, and the metallurgist the conviction that he 
does not need to go outside the products of his own country to find 
lenses of the finest obtainable quality. 

Some indication of the ability of American optical firms to produce 
photographic lenses of the highest quality may be found in the rela- 
tively new field of photogrammetry. Herein photographs taken from 
the air are used for the construction of maps. Photographs intended 
for this purpose must be characterized by the highest possible quality 
in respect of definition and by a freedom from distortion that would 
have been regarded as fantastic a few years ago. The reasons for 
these requirements are easy to comprehend. Because of unsteady 
air a flight altitude of less than a mile is impracticable. Other con- 
siderations lead to actual flight altitudes lying generally between 10,- 
000 and 20,000 feet. Photographing the surface of the earth from 
such distances, details such as road intersections, buildings, boulders, 
clumps of bushes, etc., that might serve as identifiable points whose 
location must be identified on the map must be clearly recognizable 
in the photograph. This calls for exquisite definition. On the other 


hand, the scale of the photograph must be uniform over its entire 
area. If the scale of reduction varies over the photograph measure- 
ment of distances is accompanied by corresponding errors. This 
leads to the requirement of freedom from distortion. Since perfec- 
tion in this world is unattainable, the map makers do concede to the 
lens manufacturers a little departure from the ideal an error of one 
part in four or five thousand which would correspond to an error in 
planimetry of roughly a foot in a mile. 

Aerial photographs are used not only for planimetry, however, but 
also 'or topographic mapping by employing the stereoscopic parallax 
obtained by photographing the same view from two separate points 
in the flight of the plane. There are various methods by which this 
stereoscopic parallax can be employed to yield information as to 
dimensions perpendicular to the surface of the earth, one of the most 
interesting of which is to employ two projectors to reproject a stereo- 
scopic model of that section of the earth's surface common to two 
consecutive photographs. Such projectors permit the reconstruction 
of the earth's surface in a very much reduced scale, say, one to ten 
thousand, for example, in which both horizontal distances are re- 
produced in the same scale and are measurable with a high order of 
accuracy. On such a scale a difference in elevation of five feet would 
appear as 0.15 mm or 0.006 inch. Such a value is perfectly easy to 
read so that in the stereoscopic model reproduced by such projectors 
from two pictures taken with a lens of 6-inch focus from an altitude of 
10,000 or 12,000 feet, two miles or more, it can easily be observed that 
an automobile has a definite height and is not just a flat, dark spot 
on the ground. Such results, however, can not be accomplished with- 
out lenses and mechanical work of the most exquisitely accurate 
nature. This equipment is being produced by the Bausch & Lomb 
Optical Co. in Rochester and lenses adequate for the photography are 
produced by the same company as well as by the Hawk-Eye Division 
of the Eastman Kodak Co. and the American Goerz Co. 

The drive for accuracy in this work has led to some progress in the 
field of lens testing. Too much, in the 'past, lens performance has 
been judged by purely qualitative standards. Two photographs are 
taken under what it is hoped are identical conditions and then judg- 
ment is rendered by comparison as to which is the better. In photo- 
grammetry something more has been done. Definite standards have 
been set up for definition in terms of resolving power and for distor- 
tion. The measurement of distortion is a relatively simple problem 

432 W. B. RAYTON [j. s. M. p. E. 

fundamentally although it becomes cjimcult enough in execution when 
the degree of precision required reaches the values involved hi photo- 
grammetry but the matter of definition is not so simple. The test for 
resolving power is open to criticism not for giving false information 
about a lens but because it perhaps does not give enough. Improve- 
ments are likely to follow and out of the requirements of this science 
may come some methods of evaluating lens performance that may 
more truly indicate the relative merit of lenses than the best judgment 
of the photographer no matter how conscientious he may be in reach- 
ing his conclusions. 

One thing I wish especially to emphasize and that is that whenever 
quantitative specifications have been set up for lenses we have been 
able to meet them in this country if they have been met anywhere. 

Having thus sketchily indicated whence we have come and, all too 
briefly to do justice to the subject, outlined where we stand, it is 
logical to ask, where are we going from here? 

A complete optical industry involves four things: first, adequate ', 
raw materials ; second, competent scientific research and engineering j 
application; third, skilled craftsmen to convert the products of the 
research laboratories and the engineering departments into practical , 
lenses and instruments ; and, fourth, a market for the product. 

The first of these has been discussed. For the second, we have j 
resources we did not possess twenty-five years ago. We have an 
Optical Society of America, organized twenty-four years ago, that is 
stimulating interest in both research in the field of optics and in the 
use of optical instruments and methods in the laboratory, in educa- 
tion, and in industry. We have several colleges and universities 
wherein more interest is being taken in optics and moie competent , 
instruction is available than could have been had anywhere in the | 
country twenty-five years ago. At the University of Rochester is an ; 
Institute of Optics organized to run parallel with and in closest co- i 
operation with the Department of Physics with a four-year course 
leading to a bachelor of science degree. From personal experience : 
with the graduates of this Institute I can say that they leave it with 
a knowledge of optics and of optical instruments that was formerly ; 
attainable only by years of work in an optical plant. With the incen- j 
tive set by the Optical Society and the educational advantages now j 
available, the supply of competent optical engineers, if not yet fully 
equal to the needs of the country, is certainly in a vastly better 
state than it has ever been before. 


In respect of the third point, perhaps conditions are here more fa- 
vorable also than they have been for some time. In these days of un- 
employment it would seem easy to get all the factory help that could 
be desired. We found in 1937 that such was far from the truth, an 
experience that was shared by every other industry that depended on 
| skilled workmen. It takes high qualities of intelligence, initiative, 
muscular coordination, ambition, industry, and the ability to cooper- 
ate with others to make a competent instrument maker. For a long 
time youths with such qualifications sought in a college education a 
passport to a white-collar job and disdained consideration of any job 
that involved working with their hands. With college graduates 
without outstanding qualifications for some particular kind of work 
rating at about a dime per dozen, many of them are sensibly passing 
up the $30 per week office jobs for the $60 per week of the skilled 
mechanic and they are making white-collar jobs out of such work. It 
seems not unlikely that there may be a definite movement back to the 
realization that a comfortable living obtained as a skilled craftsman 
is not an undesirable way of life for even a college graduate. 

Finally, as for the market for the product of the optical industry, it 
is increasing constantly. It is perhaps beyond reason that a product 
that practically never wears out, such as most of the products of this 
industry, should not more often develop a saturated market. Such 
things happen to a certain extent, of course, but less frequently than 
one would expect. 

The last twenty-five years, therefore, appear to have brought the 
optical industry of this country to a point where it can supply prac- 
tically anything of an optical nature that can be required. It is not 
an unimportant matter in these times, for optical equipment is a vital 
necessity for national defense in both a military and a medical sense. 
Optical instruments are indispensable tools in practically every branch 
of science and are steadily becoming more and more important in 
such fields as metallurgy, paper-making, machine shop practice, the 
manufacture of chemicals, and the manufacture of textiles. 

In this brief summary of the state of the optical art in America I 
have deliberately tried to avoid making disparaging comparisons with 
the progress of other countries. In certain lespects it is the easiest 
course to measure our progress by comparison with developments in 
other countries, and where necessary, that has been done to outline 
our progress with our needs. I believe that our present optical 
industry is completely adequate to those needs. 



Summary. A brief survey is made of the history of motion pictures in education 
and their use as a teaching aid; the present status of 16-mm sound pictures and pro- 
jectors in schools; types of educational pictures available and their sources; the dis- 
tinction between auditorium and classroom films; types and sources of auditorium 
films currently used; types and sources of classroom films; cost of films; their ac- 
ceptability by teacher groups and cooperation of non-profit educational groups toward 
creation of adequate classroom libraries; and a plan is suggested for an educational- 
film-producing organization to meet school needs. 

It is only within recent years that educators have begun to make 
use of modern aids to improve the teaching process. For thousands 
of years education depended principally on the teacher talking and 
the pupil listening. The educational process was carried on by word 
of mouth, and was strongly colored by the temperament and person- 
ality of the teacher. With the invention of the printing press, the 
scope of education became enlarged because it brought to the pupil a 
world outside the curricular descriptions of the teacher. The printed 
word brought a vast storehouse of accumulated knowledge to the 
pupil, and he became exposed to the thinking processes of many minds. 
Illustrations accompanying printed texts aided greatly in clarifying 
words. A vast world came under the observation of the pupil which 
would have been otherwise impossible, except within the limits of his 
own observations. The importance of the illustration to the teach- 
ing process is aptly described in the words of an old Chinese philoso- 
pher: "One picture is worth a thousand words." 

The printed word and the picture recorded the experiences of 
others. There still remained the element of personal experience, 
and out of this need was evolved the laboratory. Here the pupil 
could see for himself and complement his learning process by doing 
things. The laboratory is the proving ground where trial and often 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 13, 

** Ampro Corporation, Chicago, 111. 



error supply the convincing proof of the recorded laws of man and 

As the horizon of human knowledge widened, the records of the 
accumulated knowledge of mankind multiplied. Gradually, the 
teacher became a guide to the pupil, rather than the center about 
whom the entire learning process revolved. His role became more 
and more that of a leader through the complex structure of knowledge 
which was being erected. Numerous aids were developed to facilitate 
the effectiveness of teaching. Essentially, however, the three funda- 
mental processes of education still remain, listening, reading, and 

It is the seeing process with which we are here principally con- 
cerned. Visual education, as it is termed, is today one of the most 
important forces in education. It surmounts time and space. It 
extends into the past and projects into the future. It pushes out the 
walls of the classroom to the farthermost corners of the universe. 
It replaces and often surpasses the most elaborate clinical and labora- 
tory facilities. 

Motion Pictures as a Teaching Aid. Almost since the birth of 
motion pictures, educators have been aware of its value as an instruc- 
tional tool of great power and its potential value as an aid in the 
teaching process. With motion pictures, subject matter can be pre- 
sented with added emphasis to compel interest on the part of pupils. 
The use of motion makes possible more comprehensive illustrations 
supporting text-book matter and oral teaching. Animated drawings 
serve to x-ray invisible phenomena. The addition of sound still 
further enhances the effectiveness of this modern teaching tool. 

Research by Educators. Repeated experiments by leading educa- 
tors have proved definitely that pupils who have the advantage of 
the motion picture learn more and remember more. Examinations 
given to different classes reveal that those who have been aided by the 
motion picture have materially higher grades and acquire a greater 
comprehension of the subject taught than those who were not so 
aided. This makes possible an economy in time of learning which is 
of great importance to education. More important is the enrich- 
ment which comes through viewing scenes or activities which are 
exceedingly difficult or impossible to show under ordinary schoolroom 

For example, in the field of botany animated drawings can show 
plant reproduction through its entire process; stop-time-lapse pho tog- 

436 A. SHAPIRO [j. s. M. P. E. 

raphy can show the evolution and % growth of a plant; animated 
pictures can show how an automobile engine works; countries and 
peoples distant from one's own land can be brought home to the 
pupil; mechanical and electrical apparatus not commonly found in 
school laboratories can be demonstrated in operating conditions. 

History of Barriers to School Use. When motion pictures were 
first introduced to schools some seventeen years ago, it was necessary 
to use the same film equipment as was used in motion picture thea- 
ters. This meant handling 35-mm film with its well known fire 
hazards. It involved large and expensive equipment. In most 
cases a professional operator skilled in the technical complexities of 
the equipment was required. These obstacles proved a serious 
barrier to the general use of this medium and made it impracticable 
for use in classrooms. Consequently, motion pictures were confined 
to educational showings in school auditoriums before large assembled 

With the introduction of 16-mm safety film, which had first been 
developed for amateur use, these obstacles disappeared. The new 
film did not present a fire hazard, the equipment required to project 
pictures was nominal hi cost, and a professional operator was not 
required. It became practicable to use motion pictures in the class- 
room, the projector in most cases being operated by the teacher. 
Libraries of teaching films were developed, and the educator was 
given a new aid for facilitating pupil understanding. 

Up until 1929, which marked the point of general acceptance of the 
talking film, the 16-mm silent film had been making steady progress 
as an aid to education. Eastman Talking Films, with its more than 
two hundred subjects, was one of the first companies to provide a 
fairly comprehensive educational film library. The Yale Chronicles 
of America produced fifty-two reels on American history for school 
use. Hundreds of other productions were made, and distributing 
centers were formed to provide for their circulation. Over thirty 
universities developed such circulating libraries for the benefit of the 
schools in their districts. 

Introduction of Sound Pictures to Education. With the general 
acceptance of the talking film in entertainment, educators were faced 
with the problem of whether the silent or the sound-film was the best 
instrument for school use. It was quickly realized that the sound- 
film had tremendous advantages over the silent. However, once 
more there appeared a number of new obstacles. First, there were 


no libraries of educational sound-films available. Second, 16-mm 
sound-on-film was not yet developed and, consequently, the use of 
sound-film would mean going back to the cumbersome and expensive 
35-mm equipment. Third, such equipment brought with it the old 
disadvantages of fire hazard and the complexity requiring a profes- 
sional operator. On the other hand, there were so many advantages 
which the sound picture had over the silent, that the production 
of educational silent films practically ceased and much effort was 
concentrated on the problem of making sound pictures practicable 
for classroom use. 

One of the first efforts in this direction was made by Electrical 
Research Products, Inc., which produced several educational pictures 
on 16-mm film with a phonograph disk accompaniment for sound. 
In order to synchronize the picture with the sound, a very cumber- 
some and complicated equipment was developed, involving a turn- 
table attached by gears to a projector. Occasionally the needle 
would slide off the sound record and the ensuing sound would be out 
of synchronism with the picture being shown. Records would be 
misplaced and considerable care had to be exercised to insure using 
the proper record for the proper film. If a portion of the film were 
destroyed, a piece of film having the exact number of frames would 
have to be inserted where the break took place; otherwise, the ac- 
companying sound would not match the picture for the entire re- 
mainder of the reel. Altogether, it was an impracticable means for 
making use of sound pictures in the classroom. It served, however, 
to show conclusively that the sound picture was greatly superior to 
the silent for teaching purposes, and pointed the direction for future 

Available Projection Equipment. In 1930 the first 16-mm sound- 
on-film projector was publicly demonstrated. Based on present 
standards of quality, the results were very poor, but it formed a 
starting point from which the modern 16-mm sound projector has 
been developed. By 1935 the improvements had reached a state 
which led to the general acceptance of 16-mm sound-on-film for 
non-theatrical purposes. The perfection of equipment led to in- 
struments which were inexpensive as to first cost and operation, safe 
and simple to use, and capable of highly satisfactory performances. 
These improvements made entirely practicable the classroom use of 
sound pictures. 

The Present Status of Sound Pictures in Schools. In the early 

438 A. SHAPIRO [j. s. M. P. E. 

days of development of sound pictures in schools there was consider- 
able opposition on the part of teachers, based on the theory that the 
sound picture might replace many of them. It was argued that by 
bringing noted educators to the screen, classes could be enlarged and 
a good deal of the teaching processes could be carried on without per- 
sonal instruction. This, however, has not been proved to be the 
case. It is now fairly established that the sound picture is an aid, 
and in no way a substitute, for the teacher. It is another tool, albeit 
a very powerful one, to aid the teacher in the educational process, 
and today such opposition to its use by teachers is virtually unknown. 

Types of Educational Pictures. There are two general types of 
educational pictures. One group consists of pictures which are of a 
general nature containing material which correlates classroom activ- 
ity with the life outside the school. The classroom is a little world 
all by itself, and when it is devoted to the teaching of a particular 
subject or form, it tends to become isolated from the real world in 
which we live. This type of picture is designed to coordinate the 
little world of the classroom with the larger world of which it is a part. 
Pictures of this sort, being of a general nature, can be shown to 
groups of classes assembled in the school auditorium. They are, 
therefore, referred to as auditorium pictures. 

The second type consists of pictures integrated with the text-book 
and oral technic used in teaching a particular subject. These pic- 
tures may be considered as explanatory of text-book material and 
can be viewed as illustrations in motion, transcending the limitations 
of the still picture in the text-book. The sound accompanying 
these pictures creates an added effectiveness and gives continuity to 
the subject matter. This type of picture is generally referred to as a 
classroom picture, because its greatest use is in the classroom under 
the coordination of the teacher. 

This classification of educational pictures must be taken in a broad 
sense. It is often very difficult to draw the line between what is 
specifically educative and what is generally educative. Auditorium 
pictures are often effectively used for classroom teaching, and vice 
versa. The definition does not relate specifically to the place in which 
the picture is being shown. It is rather a distinction as to the place 
which the picture has in the general educational program. 

Auditorium Sound- Films. There is available for school use today 
a large supply of films of the auditorium type. These films have been 
prepared mainly from the following sources : 


(1) Theatrical pictures 

(2) Commercial pictures 
(5) U. S. Government films 
(4) Documentary films 

(7) Theatrical pictures offer the largest present source of educa- 
tional pictures for auditorium use. The vast amount of material in 
the vaults of the theatrical film makers are of significant educational 
value. By proper editing this material is being made available for 
school use. In December, 1937, the Motion Picture Project of the 
American Council of Education received a grant of $135,000 to 
finance a research program involving the evaluating and cataloguing 
of existing theatrical films so that they might be fitted into a broad 
educational program. The Motion Picture Producers and Distribu- 
tors Organization added $50,000 to this fund, thereby not only aiding 
in financing it, but giving this project the official endorsement of the 
producer groups. Approximately 8500 films of educational signifi- 
cance have already been listed. Unquestionably, this project will 
form the basis of an extensive library of films of this type. 

(2) Commercial pictures, although primarily produced for the 
benefit of their sponsors, have a definite educational value in bringing 
to the pupils information on the manufacture, nature, and use of prod- 
ucts or services. They serve to acquaint the pupils with the applica- 
tions to which their schooling serves to prepare them. Young folk 
with vague ideas about their careers are able to visualize more clearly 
what is to be expected of their future activities. These pictures 
form an invaluable background for correlating the outside world 
with the school. These pictures, incidentally, are available for 
school use in most cases without cost to the schools. At the present 
tune there are 143 commercial organizations listed as offering such 
films for school use. 

(3) The United States Government has produced more than 500 
federal films, nearly all of which have educational value. Many of 
them have been made for specific purposes, relating to the safety and 
well-being of our people. Three Federal Departments are actively 
engaged in production with their own laboratory and filming facilities. 
These are the Department of the Interior, the Department of Agri- 
culture, and the War Department. The films made by the Depart- 
ment of the Interior cover a wide variety of interesting subjects 
such as the National Parks; State and County Parks; the position 
of the Indian in this modern age; federal projects, such as the Boulder 

440 A. SHAPIRO [j. s. M. P. E. 

and Grand Coulee Dams, showing their relations to desert reclama- 
tion and their generation of power; the conservation of our oil and 
range resources; mining; pictures on our outlying possessions, such 
as Alaska, Hawaii, Puerto Rico, and the Virgin Islands; activities 
of the Civilian Conservation Corps; etc. These pictures are designed 
to illustrate the principle of the conservation of our natural resources 
and the waste resulting from the lack of such conservation. The 
Department of Agriculture has produced about 350 films covering 
the improvement of American farming. Pictures produced by the 
War Department are primarily for war training purposes. With 
the new program of rearmament in America these pictures will, no 
doubt, find use in the R. O. T. C. and other military divisions of our 

(4) Documentary films form a rather new category in the library 
of visual aids. These are films which throw light on current activities 
of sociological, historical, and political interest. An example of this 
type of film is the pictures called The March of Time. These show 
reenactments of current events which have important historical 
significance. Recently, the U. S. Government has produced several 
films of this nature. Two of these films, The River and The Plow 
That Broke the Plain, are now in circulation and have been widely 
acclaimed by educators as outstanding productions hi this field. 
The approbation which these two pictures have received from educa- 
tors will undoubtedly stimulate the production of more pictures of 
this character. 

The Classroom Sound- Films. Present methods of teaching do not 
permit very much time to be spent in auditorium group meetings 
correlating school activity with outside interest. The classroom 
forms the foundation of our educational system and it is here that the 
greatest lack of sound-film exists. Up to the present, the principal 
source of this material has been Erpi Pictures Consultants, Inc., 
which has produced about 110 classroom sound-films. Unfortu- 
nately, most of these films have been designed for use in colleges and 
high-schools. The need for films for grade schools, and even the 
kindergartens, is only now beginning to receive some belated atten- 

There is a great immediate need for the production of a compre- 
hensive library of classroom sound-films. It has been estimated 
that to provide sound-film pictures to accompany grade school in- 
struction, approximately 300 films are required. For high schools, 


at least 200 films are necessary to complement text-book teaching. 
What is now available does not make even a beginning of such a 
program and has merely served to indicate the possibilities for better 
teaching which the sound picture promises. 

It is quite likely that after classroom pictures have become widely 
used, such use will have a profound effect upon the character of the 
text-books used. It is even conceivable that text-books will be 
written with the idea of coordinating the sound pictures with the 
text to the best advantage. For the present, however, the classroom 
pictures can be so designed as to blend with the majority of the text- 
books used for a particular course. The pictures would help clarify 
and make the text-book more readily comprehensible. They would 
form an invaluable complement to the course of study. 

For most effective use the classroom picture should be timed so as 
to fit in with the program of the subject being taught. The effective- 
ness of the film is very much decreased if it is not shown at the time 
the subject is under discussion. It is then that it can create its most 
effective impression on the pupils. For this reason, it is believed that 
wherever possible each school should have its own library of class- 
room films. In this way the films are available whenever they are 
needed and are not subject to the inevitable delays to which any 
circulating plan is subject. 

Cost of Films. At present very few schools maintain their own 
libraries, but depend chiefly upon the rental of films from distribut- 
ing organizations maintaining circulating libraries of films. Some 
thirty-nine universities as well as numerous state or local departments 
of education are engaged in this activity. One of the chief factors for 
the development of these distributing organizations is the high cost 
of films, which has been a serious barrier to the purchase of school 
libraries. It has been proposed by mass production to reduce the 
cost of films so as to make possible a package-plan low enough in cost 
to permit schools to purchase their own libraries of classroom films. 
It has been found by experience that the use of classroom films is 
restricted according to the degree in which the films are available 
and, by making the films available at all times to its classes, the indi- 
vidual school can obtain the greatest benefit from them. 

It is generally understood that the cost of making present class- 
room films has varied from $4000 to $10,000 per subject. Since 
these subjects have been produced in the past under somewhat ex- 
perimental conditions, this cost can be considered much higher than 

442 A. SHAPIRO [j. s. M. p. E. 

would result in a program involving planned economy. Only a 
small number of prints have been made of each subject, so that the 
per print production costs have been the largest factor of film cost. 
A plan assuring an outlet for a large number of prints of each subject 
would reduce this cost substantially and make it possible to place a 
low sales price on such prints. 

If from 3 to 5 per cent of the total of 276,000 schools may be thus 
served, approximately 10,000 prints of each subject would be re- 
quired. On this basis, the production cost per subject would be $1 
or less per print, and the total cost, including film and printing, would 
be between $5 and $6 per print. This should make possible a sale 
price of about $20 per subject, which is less than one-half the cost of 
the present sale price of corresponding subjects. It is believed that 
a distribution of 10,000 prints per subject is a very conservative esti- 
mate of the potentialities of this field. Already some 7500 schools 
possess 16-mm sound equipment. It would need only the promise of 
such a library to multiply this figure many tunes. 

Acceptibility. Motion pictures in education have captured the 
imagination of the public and of educators. Active cooperation is 
being given by leading educators and educational organizations for 
the use of subject matter and the best methods of presenting it. 
Teachers who are now using motion pictures are enthusiastic about 
the results obtained and render all possible cooperation. A number 
of colleges are providing courses for teachers in the use of equipment 
and film. The National Educational Association, with 157,000 
teacher-members, has a special division devoted to motion pictures 
in education. In very few fields of endeavor is as much cooperation 
available so freely and so sincerely as in the field of visual education. 

Cooperation. Foundation groups such as the Rockefeller Founda- 
tion and the Carnegie Institute have made available funds for re- 
search and aid in furthering the use and production of motion pic- 
tures for schools. The General Education Board of the Rockefeller 
Foundation has a motion picture project involving the evaluation of 
existing films as educational aids. Another Rockefeller Foundation 
called The Association of School Films Libraries, Inc., is engaged 
in the work of compiling a catalogue covering these evaluations. The 
Commission on Human Relations of the Progressive Films Associa- 
tion is investigating the work of reediting films of theatrical dis- 
tributors for school use. The Museum of Modern Arts Film Library 
is collecting a historical library of films. The American Film Center 


acts as adviser to film producers who engage in making pictures for 
the educational field. All these are non-profit organizations operat- 
ing by means of private funds for public benefit. 

Due to its serious purpose, the motion picture in education can not 
be treated in the same manner as a picture made for entertainment. 
First and foremost, it must be considered as a pedagogic tool and the 
art of education becomes more important than the art of film produc- 
tion. Most assuredly, the best equipment and technical facilities 
must be at the disposal of the producers of educational motion pic- 
tures. But the guidance and the correlation with school curricula 
are best controlled by educators. In a sense, the production of 
educational pictures represents a new art which is a blend of the best 
of the older technic with modern science. 

Suggested Organization of Educational Film Production. There is 
a definite place at present for an organization which will coordinate 
all the findings of the various research groups and devote itself to 
making the educational pictures which are most needed. While the 
profit motive need not be ignored, it is highly desirable that the 
principles of public interest be given first place. A suggested pro- 
gram for such an organization is as follows : 

(1) Form an independent council of leading educators to act in an advisory 

(2) Prepare surveys of subjects taught and text-books used, and short synopses 
of proposed appropriate films. 

(5) Develop scenarios under expert scenario-writing supervision, coupled with 
educator editing. 

(4) Arrange for suitable production facilities. 

(5) Appoint publicity director to handle school and public relations. 
(0) Prepare sales and exploitation plans with appropriate personnel. 

(7) Coordinate efforts of production, sales and exploitation together with 
advisory council. 

The Social Aspect. It is felt that the field of supplying educational 
pictures offers not only an opportunity for profit, but also for public 
welfare work of a high degree. The educational system of America 
is the foundation of our system of democracy. Any improvements 
therein are bound to have fundamental effects upon the character of 
our future citizens. The possibilities for good effects are boundless 
and well merit the enterprise and support of those who are socially 


Summary. The color intensity of a motion picture projection system may affect 
the presentation of the picture. Using sample comparison methods the average 
color of projection systems can be determined and the deviation of a particular system 

Considerable data have been collected and published concerning 
motion picture screen illumination from the standpoint of intensity. 
There is very little information concerning the color of projection 
systems. A projection system includes the light-source, optical 
system, and screen. The color of the light-source may vary from 
the yellow of a Mazda lamp to the blue-white of a high-intensity 
arc. The quality and condition of the elements of the optical 
system will affect the color of the light leaving the projector. The 
color of the light reflected from the screen will be affected by its type 
and condition. 

A rapid and simple comparison method of determining the color 
of the system by measurement of the light reflected from the screen 
has been developed. The equipment used consists essentially of a 
light-source of adjustable color and intensity and a power-supply 
unit. The variable light-source consists of a slide-film projector 
unit equipped with a 500-watt Mazda lamp, a lens with iris dia- 
phragm, and a filter holder mounted in front of the lens. The power- 
supply unit consists of an autotransformer with an output range up 
to 130 volts when supplied with 115-volt, a-c power. A voltmeter 
and ammeter are wired into the unit which also serves as a support 
for the projector. Auxiliary equipment includes a set of "Daylight 
Blue" filters ranging in thickness from 0.04 to 0.10 inch in steps of 
0.01 inch, a white "standard screen," and a set of colored reflection 
screens. The technic of making a screen-color measurement with 
the "comparator" is essentially as follows: 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received 
April 18, 1939. 

** Technicolor Motion Picture Corp., Hollywood, Calif. 



(1) Set up the standard screen as near the center of the projection screen as 

(2) Set up the comparator in front of the screen and adjust the beam so that 
it just covers standard screen. 

(5) Operate the projector system, and erect a mask to shield a small standard 
screen from the light-beam. 

(4) Insert a suitable filter in the comparator and adjust the intensity to 
match that of the projector system. 

(5) Refine the color match between the standard screen and the screen being 
measured by adjustment of the filters, and record the filter thickness. 

(6) An alternative method employs tinted screens instead of a white-screen- 
and-filter combination. The procedure is essentially the same. 

Interpretation of the results of comparative measurements may 
be in arbitrary units such as filter thickness, which in turn may be 
translated into other units as desired. 

A number of measurements have been made with the comparator 
in East and West Coast theaters, and in many of the Hollywood 
studios. Analysis of the data indicates that there is appreciable 
i variation in color of projection systems. The average system color 
may be reproduced with a 0.13-inch Daylight Blue filter in the 
comparator; deviations of 0.03 inch in thickness from the 0.13 
average are found. 


MR. CRABTREE: How does this variation in color affect audience reaction? 
That is the only way we can estimate the importance or seriousness of this varia- 
tion in screen color. Have you made any audience tests? 

MR. HARCUS: Not in the sense in which you are thinking, I believe. In 
viewing a picture on the screen, whether black-and-white or colored, the viewer, 
unless the color deviation is very marked, may not notice the difference in terms 
of neutral gray in the case of black and white, or the color variation in the case of 
a colored picture. If the deviation is very marked the difference becomes very 

MR. HOOPER: What kind of instrument do you use to determine the color of 
the light on the screen in case you want to get any particular color? 

MR. HARCUS: We are using this type of comparison unit with which the color 
of the screen is matched, as demonstrated. 

MR. HOOPER: Do you have any instrument with which to measure the color- 
temperature of the light on the screen? 

MR. HARCUS: We have not yet undertaken any studies to that extent. 

MR. CARLSON: Does the range of color difference shown here represent an 
acceptable range? Do you have any means of measuring screen brightness 
rather than incident light? In several instances where a balance presumably 
was demonstrated, differences in brightness seemed to be apparent. Was that 

446 W. C. HARCUS [j. s. M. p. EJ 

due possibly to the fact that the photocell is not color-corrected and due to the 
selective reflectivity of the screen itself? 

MR. HARCUS: There is the degree of difference we have observed here. I 
believe this represents a greater deviation than is desirable for the presentation 
of either black-and-white or colored pictures. The meter we use for this type of) 
work is a simple Weston meter. This is a photoelectric type of meter and mea- 
sures the light falling on the screen. It is not accurately color corrected so far 
as color-sensitivity of the eye is concerned. 

MR. KELLOGG: I take it from your answer to Mr. Crabtree's question that 
the audience is not very critical of the color and that if the color source did varyj 
over a certain range the audience would not criticize it. If that is the case, 
I do not quite see the purpose of making this rather exact study. 

For color projection do you find a still greater premium on very high screen 
brightness than you need for black-and-white pictures and do you get more return 
psychologically from high brightness in the case of color than you do black-and 

MR. HARCUS: The purpose of the initial investigation was to determine the 
average screen color of the many theaters we encounter. This was to be de- 
termined as a control for the manufacture of color pictures. Color pictures that 
look the best on the "average screen" may not look quite so good on screens that 
deviate from the average. 

We find in terms of light falling on the screen, that the average large motion j 
picture screen does not exceed ten foot-candles by any substantial amount. Wei 
have found some running as high as thirteen and a few as high as eighteen or! 
twenty. We have found some as low as five foot-candles at the center of thej 
screen. We manufacture color pictures to show to the best advantage on ani 
average screen of ten foot-candles. 

MR. RICHARDSON: In lighting for color photography we have a problem quite < 
similar to the one here under consideration. The requirements of color photog- j 
raphy rather closely limit the spectral quality of the sources used on the motion ' 
picture set. Until we devised a qualitative instrument for measuring the spec- 
tral distribution of light-sources, we had only our eyes with which to make com- ! 
parisons. In lighting for color photography, we have a very satisfactory re-j 
producible standard in the M-R Type 90 high-intensity arc lamp using a 13. 6- 1 
mm carbon operating at 120 amperes with 57 volts across the arc. 

Before we devised our comparison instrument, we had to go through an elabo- { 
rate routine of photographic testing whenever a new unit was under development, \ 
in order to adjust the arc voltage and amperage of the unit to give a proper i 
balance in the light. The instrument consists of three small General Electric j 
light meters assembled in a light-tight box, each before a window provided with j 
an adjustable shutter. Each of the windows carries a color-separation Wratten I 
filter one gr(!en, one red, one blue. We use the instrument comparatively, 
first setting the shutters to give equal readings of 50 foot-candles through thei 
three filters from our standard M-R Type 90 light-source; then readings are! 
taken on the source under examination, and a comparison is obtained. WeJ 
choose the 50-foot-candle readings as standard because the small General Elec-l 
trie light meters had been selected to read equally in this position. Having.) 
separate readings through each of the separation filters, we are able to study the i 

)ct., 1939] SCREEN COLOR 447 

ght-source under examination and explore for the conditions that will bring it 
lost nearly to the standard required. 

| We use this instrument qualitatively through the development of our Duarc, 
jnd it may be of interest that our observations made with the instrument cor- 
lesponded very well with the photographic check results made in the Technicolor 
aboratories. We do not use the instrument quantitatively, although if it were 
jiore carefully developed, it might be arranged for quantitative color measure- 
ments. The instrument could probably be developed to analyze screen bright- 
less and be of particular value for studying screens and light-sources used in the 
rejection of color motion pictures. 

MR. ZURICH: What have you found to be the best type of projection screen 
jr Technicolor work? 

MR. HARCUS: I am sorry to say I can not answer that question. We take 
he equipment just as it comes, and your question has not been part of the study. 

MR. ZURICH: If you take anything from a dirty bronze yellow to a dark blue 
n the various screens, the color will show up very well on certain screens and 
nth certain projected light and not so well under other circumstances. Surely 
jhere must be some "average" screen. 

I MR. HARCUS: A screen that is clean and white will certainly present a picture 
f any type much better than one that is off color or one that is not clean. 

MR. ZURICH: If the light from the various lamp houses ranges from a dirty 
[ronze yellow to a dark blue, in addition to having a dirty screen, that is some- 
tiing to consider. 

MR. HARCUS: That is what we are trying to simulate. We have this device, 

hich has enabled us to determine the average screen color in the average theater. 

MR. ZURICH: We t all face the need today for an instrument that will measure 
!he light reflected from the screen one that we can take into our auditoriums 
|nd measure the reflective quality of the screen as well as the color-temperature. 
?his will give us a standard that we do not have today. 

MR. HARCUS: That is right. I believe that some instruments have been 
eveloped which may soon become available. 

MR. RICHARDSON: In England there was at first a very heavy import of 
Lmerican carbons for use in Technicolor photography and projection. One 
English manufacturer has undertaken the problem of supplying the English 
(narket with suitable carbons for photographing Technicolor pictures and has 
pproached the problem in a very scientific way. They have developed an 
astrument that would almost exactly meet the need you mentioned. In England 
he air is foggy and there is considerable soot in the atmosphere. Screens de- 
eriorate quite rapidly. These English carbon manufacturers have developed 
,n instrument for installation in the projection booth which integrates the screen 
llumination and enables the projectionist at periodic intervals to check his screen 
nd each projection lamp against the optimum performance that was established 
hen the screen was new or was re-surfaced. This instrument would meet the 
equirements mentioned by Mr. Zurich. 

MR. JOY: In regard to this instrument that Mr. Richardson referred to, I 
>elieve that it does give some indication of the total relative light on the screen 
iut it does not take into consideration the distribution of the light, that is, the 
elation of the light at the sides to the light at the center or the color of the light. 

448 W. C. HARCUS [j. s. M. p. E. 

One of the things we have found time and time again in our own tests is that if 
we do not take into consideration the distribution of light, the total light read- 
ing is often misleading and does not necessarily give a true comparison. 

MR. GRIFFIN: The Projection Practice Committee is trying to get an in- 
strument that will indicate when the screens are deteriorating. We are nol 
primarily interested, for this purpose, in an instrument that will give the light 
distribution. We simply want to be able to determine what the brightness is 
and when the screen needs changing or resurfacing. We have not been able tc 
find any such instruments other than illuminometers like the Macbeth, when 
one must read the brilliancy of a target or other surface. Instruments of thi; 
type are expensive, and apart from that it is difficult to get two men to agree orj 
identical readings. We want an instrument that is definite in results, simple t( 
operate, and with which comparisons can be made from time to time, beginning 
when screens are first installed and periodically thereafter. 

MR. RICHARDSON: The English instrument referred to is designed especialh 
for that, and is intended to be used in the projection room. In England there 
are fewer of the small theaters which we designate as neighborhood theaters, o 
which we have so many in this country, but in their larger theaters they cai 
afford to install this instrument which gives such complete information. 

MR. JOY : Getting back to Mr. Harcus' paper, is the small comparison screeJ 
that you use a standard color or are these sheets of colored paper that you usecj 
standard color sheets? 

MR. HARCUS: It is a white target screen made from a white cellulose board 
The colored screens are manufactured to match certain screens that we have en 
countered. They vary as to their content of blue, blue-green, and magenta, am 
represent white closely when reflecting mazda light. . 

MR. GARBER: Has there been set up any standard of screen brightness. 

MR. GRIFFIN : The Projection Practice Committee, after some years of study ! 
ing the problem, have submitted a recommendation to the Standards Committee 
As a recommended practice it has been suggested that 7 to 14 foot-lamberts b 
the range of screen brightness, with no film in the projector and with the shutte 
running. It is a matter of selecting the proper illuminant behind the film an< 
the proper optics to obtain the best result. 

MR. HARCUS: The instrument that Mr. Richardson has described is essential!; 
a camera with a very sensitive photographic cell mounted in focus behind th 
lens. The cell is wired to a calibrated meter which can be mounted in front o 
the projectionist. 

MR. KELLOGG: About the desirability of high illumination, Mr. Harcus sai< 
that an effort was made to make the Technicolor film to project satisfactoril; 
with a screen illumination representing what one might expect in the averag 
good theater. Presumably that is a matter of not making the prints so dens 
that they will be too dark under those conditions. Would the same picture loa 
still better if projected with higher illumination? 

MR. HARCUS: My personal opinion is that any good print will show to bette 
advantage on a brighter screen. A print that will show to good advantage o 
a 10-foot-candle screen will show to better advantage on a brighter screen. - 
print made to show to best advantage on a 2-foot-candle screen would probabl 
be "burned up" at higher illumination. 


During the Conventions of the Society, symposiums on new motion picture ap- 
paratus are held, in which various manufacturers of equipment describe and demon- 
strate their new products and developments. Some of this equipment is described 
in the following pages; the remainder will be published in subsequent issues of the 


The trend toward mobile recording equipment has been brought about by a de- 
sire to use the equipment interchangeably for stage and location work. It is 
further emphasized by new sound-stage construction, sufficiently remote from the 
original centralized recording plant to make it uneconomical to supply such re- 
cording facilities from one point. Portable channels, 1 complete from micro- 
iphone to film recorder, have been developed to fulfill these needs. Their use has 
'demonstrated that a high degree of mobility may be had without a sacrifice in 
overall quality or operating ease. Experience gained with such equipment has 
indicated a reduction in the number of units comprising the channel, and an elimi- 
nation of facilities not requisite to the general run of production recording. As a 
result, a light-weight channel has been designed for production recording. 

The basic design requirements for the channel were formulated in conjunction 
with engineers of the Metro-Goldwyn-Mayer Sound Department, for whom the 
'first channel was intended. At the same time, sufficient flexibility was allowed, 
so that the basic design might serve as the foundation for a simplified system for 
other users. 

The complete system comprises two main units: the first, a stage or pick-up 
unit containing mixing and monitoring facilities, an amplifier with sufficient gain 
and carrying capacity to operate a light-valve, and a carrier type of noise-reduc- 
ion unit; the second, a film-recording machine containing the modulator and 
not or system controls. Power for the system may be supplied from a-c operated 
ectifiers or from batteries. The units are interconnected by means of standard 
6-conductor cables equipped with plugs and jacks. A schematic drawing showing 
the transmission features of the apparatus connected for operation is shown in 
Fig. 1. The appearance of the amplifier-noise-reduction unit is shown in Fig. 2, 
and the recorder in Fig. 3. 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 1, 

'* Electrical Research Products, Inc., Hollywood, Calif. 




^General Design Features. Close cooperation with studio personnel has strongly 
influenced the physical design and the transmisson and operating features of the 
channel. The case housing the amplifier-noise-reduction unit is duralumin, ade 
quately braced for strength and containing doors affording access to the equip- 
ment items. A chassis type of mounting is used carrying single side-mounting 
equipment units. Wiring is from point to point, and is confined to the termina: 
side of the mounting chassis. Vacuum-tubes hi low-level circuits are shock 
mounted. Shielding is employed between vacuum-tubes and equipment part; 
wherever necessary to eliminate cross- talk. Low-level transformers are magneti 
cally shielded with permalloy to reduce pick-up due to exposure to power fields 

11 filament and plate circuit metering is accomplished by means of a switch anc 

TO CI 1 









1 T 


1 ^ 



1 :>- L - : 



8fr iHir 




IOM.TOI IOBCR rf, u* crric wit FLTCR 










*>'. OK 

i* "*" 

jiP r 

FIG. 1. Transmission features of the equipment, connected for operation. 

meter. The meter indicates percentages, reading 100 per cent when operation i 

Transmission Features. Heater-type vacuum-tubes are employed throughout 
allowing operation on either a-c or d-c sources. The use of special tubes in som 
instances has made it possible to combine circuit functions, with resulting simpli 
fication in equipment. Tubes having extremely low noise and low microphoni 
outputs are used in low-level stages. All tubes are self -biased. 

All grid and plate circuits are protected by filtering, minimizing disturbance 
resulting from coupling through a common power-supply. Further improvemen 
in the amplifier's stability is secured by the application of negative feedback to it 
circuits. 2 

The maximum gain of the amplifier system is 80 db. This is somewhat les 
than that of previous systems, but is adequate since the amplifier is intended t 
operate from single-stage condenser-transmitter amplifiers, whose output is cor 
siderably higher than that of the moving-coil microphone. The latter microphon 
may be used with this channel by the introduction of a single-stage amplifier be 


tween the microphone and the recording amplifier. The frequency-response 
characteristic of the amplifier is essentially uniform from 40 to 10,000 cycles. Its 
carrying capacity is adequate for light-valve operation. 

Amplifier -Noise-Reduction Unit. All recording facilities except the film re- 
corder are contained in this unit. It is housed in a duralumin case 16 inches long, 
9 inches wide, and 12 inches high, weighing 42 pounds. All operating controls and 
meters are located on the top panel which is hinged to the case. Raising the panel 
permits access to the panel-mounted equipment, as well as to all other equipment 
units mounted on the chassis beneath it. Small doors near the base at opposite 
ends of the case cover jack connections to the microphones and film recorder. A 
door in the front of the case provides access to all vacuum-tubes. The entire base 
is easily removed by means of four thumb-screws, and gives access to the wiring 
compartment. The base is gasket-sealed, preventing moisture from entering if 

FIG. 2. The amplifier noise-reduction unit. 

the unit is set in damp places. Fig. 4 is a view of the case with the hinged control 
panel opened, exposing the chassis-mounted equipment. 

The operating controls on the top panel are of two types, knob and screw-driver 
adjusted. The former includes those used at all times; i. e., the mixer potenti- 
ometers, microphone cut-off keys, and metering switch. The latter are usually ad- 
justed only once or twice a day, and include noise-reduction margin and bias con- 
trols, monitoring volume, and volume- indicator sensitivity. Fig. 2 illustrates the 
control panel. 

The functions and characteristics of the circuit elements indicated in Fig. 1, 
exclusive of the noise-reduction, are as follows : 

(Mixer) A two-position mixer using 200-ohm step-by-step "Bridged T" type 
potentiometers is employed. Microphone cut-off keys having click suppressors 
are associated with each mixer position. 

(Main Amplifier) The mixer is transformer-coupled to a three-stage amplifier 
having a push-pull output stage. Negative feedback is applied to the first two 
stages as well as to the output stage. The 1000-cycle gain of the amplifier is 80 
db, and its frequency-response characteristic is essentially uniform from 40 to 



10,000 cycles. The output impedance is .nominally 500 ohms and is connected 
through the various equalizer and filter networks to the light-valve transformer. 
The amplifier's carrying-capacity using 180-volt B supply is +16 db/0.006 watt 
for 1 per cent total harmonics at 400 cycles. An increase in power output to meet 
special requirements may be easily secured by an increase hi plate voltage. 

(Networks') Various types of special networks may be supplied to meet particu- 
lar recording requirements. Those contained in this unit are low and high-pass 
filters, and a pre-equalizer. All networks are easily removed from the circuit by 
strap connections, which also facilitate testing the individual networks. The 
low-pass filter has a nominal cut-off at 8000 cycles and contains an Af-derived end- 
section, permitting adjustment of its peak attenuation to a frequency correspond- 
ing to the tuning frequency of the light- valve. The high-pass filter has a nominal 

FIG. 3. The recorder. 

cut-off at 50 cycles. The pre-equalizer is a constant-resistance network of the 
type having a 6-db insertion loss at 1000 cycles. 

(Power Level Indicator) The power level indicator is of the high-speed, criti- 
cally damped type. It is supplied by a single-stage amplifier bridged across the 
main amplifier output following the high and low-pass filters. A sensitivity switch 
permits adjustment in 2-db steps from - 10 db to +10 db/0.006 watt. 

(Monitoring) Monitoring is provided by means of a bridging transformer con- 
nected across the main amplifier's output between the filter and pre-equalizer. It 
is intended to supply a high-quality moving-coil headset. Volume control in 2- 
db steps over an 18-db range is provided. 

(Metering) A single meter in conjunction with a switch and metering shunts is 
used to measure vacuum-tube plate currents, plate and heater voltage, and noise- 
reduction bias current. The meter has a double scale, reading percentages for 
vacuum-tube measurements and milliamperes for bias current. 


(Power Requirements') For normal operation the power required is 3 amperes 
at 6.3 volts, and 0.054 ampere at 180 volts. 

Noise-Reduction. The noise-reduction circuit is of the carrier-modulated type. 
Referring to Fig. 1, the output of the recording amplifier is connected through the 
recorder circuits to the noise-reduction input. Here the signal is connected 
through a gain control to a single-stage amplifier. The signal is amplified with a 
slightly rising frequency characteristic (about 4 db at 8500 cycles). The signal 
is then rectified by the diode elements of a duo-diode triode vacuum-tube, filtered, 
and caused to modulate a 20-kc carrier in a manner proportional to the signal 
envelope. The modulator stage employs a single vacuum-tube grid biased be- 


FIG. 4. View of the amplifier noise-reduction unit with 
the hinged control panel open. 

yond cut-off, using grid voltage variation for controlling the transmission of the 
20-kc signal through it. The triode elements of the duo-diode triode are used in an 
oscillator stage generating a frequency of about 20 kc. After subsequent ampli- 
fication, rectification, and filtering, a bias current having a range of 0-400 mils in 
a 1-ohm load is obtained. 

The input required to cancel the bias current to zero is approximately 1 db 
relative to 0.006 watt across the 500-ohm circuit. The controls for bias current 
and bias cancellation are screw-driver adjusted so that once set they will not be 
inadvertently disturbed. The unit is set up for an operating time of 0.014 second, 
and a restoring time of 0.032 second. 

These figures represent the time required for 90 per cent of the change to occur 
when a test tone is keyed on or off, respectively. This timing is suitable either 



for normal single-track recording or for push-pull recording. Owing to the fact 
that the rectifier-filter circuit is so designed that the cancellation of bias current 
is proportional to peak signal amplitudes rather than to average amplitudes, the 
margin may be reduced considerably over that used in older equipments. This 
in turn permits reduction of the operating time to the value shown. 

Since no photocell monitor has, as yet, been supplied with this equipment, the 
necessary bias current may be determined by removing the light-valve from the re- 
corder and setting it, with magnets, on a microscope jig. After connecting the 
light-valve terminals to the amplifier-noise-reduction unit, the closure current 
may be determined by direct observation of the ribbon deflection. The light- 

FIG. 5. The recorder, with part of the duralumin case removed. 

valve overload point may be observed in a similar manner when a constant fre- 
quency is impressed, or other methods may be employed, depending upon the de- 
gree of precision desired. 

Film Recorder. The chief consideration in the design of the film recorder was 
to keep the weight at a minimum. The problem was complicated by the require- 
ment that all the motor and modulator control switches, film-speed indicator, 
film-footage counter, and exciting-lamp current meter, were to be mounted as 
integral parts of the film recorder. 

As the basic design arrangement of the film-pulling mechanism of a preceding 
type of recorder known as the RA -1007 -A l had proved practicable as regards 
weight and sound-quality, it was decided to incorporate a similar film-pulling 
mechanism in the new type. Instead of casting the film-pulling compartment, 


gear- assembly housing, and base in one piece, the base was cast as a separate unit, 
to which the recorder case proper can be attached later. The base which serves 
mainly as a frame for mounting the sundry switches, meters, and other necessary 
equipment used in operating the recorder, can be made light in weight. The re- 
corder case being nearly cubical in exterior dimensions about 20 by 13 by 13 
inches allows unusual resistance against mechanical warpage, which, when pres- 
ent, results in poor alignment of shafts and gears. As a further aid in reducing 
total weight, sheet duralumin was used wherever possible. These precautions 
enabled the total recorder weight to be kept to 100 pounds. 

Fig. 5 shows the recorder assembled and mounted on its base with the duralumin 
parts removed, thereby exposing the driving motor, modulator, and other associ- 
ated equipment. Fig. 3 shows the same recorder, except for the film magazine, 
completely assembled for operation. 

(Controls) Referring to Figs. 3 and 5, the controls incorporated in the recorder 
may be examined. At the extreme left is the film-footage counter. It is driven 
by means of a flexible shaft, the driving end being coupled to a gear assembly 
driven directly from the end of the driving-motor shaft. To the right of the foot- 
age counter are four electrical switches for controlling the modulator speech cir- 
cuits, the exciter-lamp circuit, and an overall start and stop switch for the recorder 
and interlocked camera-driving motors. Below these four switches are a monitor- 
ing jack and a lamp-current rheostat knob. The meter in the center of the panel 
indicates the exciter-lamp current. 

To the right of the lamp -current meter are three sets of motor-control switches. 
This type of channel is so arranged that either one or two picture camera motors 
can be interlocked to the recording motor. All the switches and field rheostats 
which operate these three motors are entirely under control of the recorder man. 
Thus, each set of switches, together with the rheostat control knob immediately 
below each set of switches, provides independent switching facilities for each 
motor, whereas the master start switch located immediately to the left of the lamp- 
current ammeter controls any motors which have been previously connected to 
the interlock circuit. The meter at the extreme right is a film-speed indicator of 
the vibrating-reed type. It is driven by a-c power derived from one of the three 
phases of the interlock circuit. As each motor can be independently connected 
to the interlock and d-c power-circuit, it is possible to use this speed-indicator to 
set the average speed of every motor before interlocking. Such a presetting 
to about proper operating speed for each motor results in a minimum of strain 
on the interlock circuit, allowing quick and reliable starting of the entire inter- 
locked motor system. 

Fig. 6 shows the under side of the base, exposing the miscellaneous wiring, re- 
sistances, and other components. The large vacant space below the recorder case 
and toward the rear is for use in housing a two-stage PEC amplifier when such a 
facility is desired. 

(Modulator) The modulator, shown in Fig. 5, uses the latest type of permanent- 
magnet light-valve, and employs some new design features. In order to allow the 
use of an objective lens system covering 200 mils, as demanded by some studios 
for push-pull recording, a longer distance from light -valve to film was necessary 
in order to provide sound-quality comparable to that of the fixed-channel studio 
recorders. This longer optical path distance, due to minimum space require- 


ments, necessitated bending the light from the exciter lamp through an angle of 90 
degrees before passing into the condensing lens of the modulator. While no 
PEC monitoring facilities were required by the studio for which the recorder was 
designed, additional optical parts may be added to the modulator to provide this 
facility. A visual monitoring means is incorporated, obtained by deflecting a 
small amount of light from the light-valve to the lens path and focusing the light- 
valve ribbons by means of lenses and prisms on a rectangular ground-glass screen. 
This screen is shown in the lower left side of the modulator. The image of the 
ribbons on the screen is enlarged about two to one, allowing easy inspection of the 
light-valve ribbon edges at all times. This visual monitoring feature, although 
primarily incorporated as a substitute for PEC monitoring, need not be discarded 
when PEC monitoring is provided. The exciting lamp is of the 9-ampere, 11.1- 
volt type, in order to provide ample coverage over the entire length of the light- 

FIG. 6. Base view of the recorder. 

valve ribbons. In this film-recorder electrical power for the exciter-lamp is 
obtained from a generator winding which is a part of the driving motor. If de- 
sired, this feature may be omitted, and the lamp current obtained from a 12- volt 
power-supply. The lamp is mounted in an adjustable socket which allows simple 
and ample adjustment of the lamp in any position without the use of any tools. 

(Motor System) The driving motors for this particular channel are designed 
to operate from a 120-volt, d-c power-supply instead of from the customary 12-volt 
source. This higher line voltage results in two advantages : first, the line-current 
being considerably reduced, line-drop troubles are eliminated at the camera 
motors; second, the low line-current makes it practicable to supply all the motors 
from a common battery power-supply located at the film recorder, and do all the 
switching and controlling at the recorder position. 

The motor system for the channel is of the d-c interlock type ; that is, the motors 
are essentially d-c. operated, the 3-phase interlock circuit being obtained by tap- 
ping into the d-c. commutator. In order to prevent direct short-circuits in the 
interlock circuit, two sets of three ballast-lamps are used between the recorder and 


each of the camera motors. These ballast-lamps have the characteristic of having 
a very low filament resistance when cold (low current flow), which rises to a 
fairly high value when hot (high current flow). Thus, on a direct short-circuit 
which ordinarily causes an excessive interlock current flow, the corresponding 
ballast-lamp resistance rise introduces an effective curb to limit the total current 
flowing in the interlock circuit. At normal interlock conditions, when the current 
flow is relatively low, the lamp resistance is a minimum, and therefore produces 
no appreciable effect in the interlock circuit. 

The speed of each motor when not connected to the interlock circuit is control- 
lable by variation of its associated field rheostat. These rheostats are all located 
in the base of the recorder, connecting leads running out from the recorder through 
the motor cables to the camera motors. The operating speed for the channel de- 
scribed in this paper is 2880 rpm, but other speeds such as 3600 or 1800 rpm can 
be obtained by a slight change in motor winding constants. If one of the latter 
speeds is used it is possible to interlock the channel to the standard 3-phase, 60- 
cycle commercial power system and thus obtain very excellent speed control when- 
ever regulated 3-phase power is available. 

The light-weight recording system which has been described has been used for 
considerable studio production. The results of this operating experience have 
shown that the system is capable of a standard of performance comparable with 
that of the more elaborate studio channels. The added convenience of the few 
equipment units emphasizes again the trend toward this type of recording system. 


Quality Portable Film-Recording System," /. Soc. Mot. Pict. Eng., XXVIII 
(Feb., 1937), p. 191. 

2 BLACK, H. S.: "Stabilized Feedback Amplifiers," Bell Syst. Tech. J., XIII 
(Jan., 1934), p. 114. 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C. 

American Cinematographer 

20 (July, 1939), No. 7 
British Cinematographer Talks of Hollywood (pp. 303, F. YOUNG 


Shoot Three Dimension Picture with Polaroid (p. 304) 
Time and Temperature Versus Test for Negative (pp. I. MILLARD 


Dye Transfer Enters Commercial Field (pp. 310, 327) 
Doolittle Builds Rewind and Film Viewer (pp. 312-314) W. STULL 
Kodak Puts on Market Its Supermatic Shutter (p. 318) 

20 (August, 1939), No. 8 
White Border Screen? (pp. 344-345) 
Debrie Building Rugged 16 Mm. Reduction Printer R. F. MITCHELL 

(pp. 348-349) 

Faster Color Film Cuts Light in Half (pp. 355-356) 
Process Shots Aided by Triple Projector (pp. 363-366, 


British Journal of Photography 

86 (June 30, 1939), No. 4130 
Progress in Colour (pp. 405-506) 

86 (July 28, 1939), No. 4134 
Progress in Colour (pp. 467-468) 


19 (July, 1939), No. 7 

Sound Motion Picture Films in Television. Ill (pp. 

Fernseh, A.G. 

1 (July, 1939), No. 4 

Zehn Jahre Fernsehtechnik (pp. 111-122) R. MOLLER AND 





The Ideal Kinema 

7 (July 20, 1939), No. 80 
Imperfections in Projector Design (pp. 29-30) 

International Photographer 

11 (July, 1939), No. 6 
Projection Symposium, Parts VII, VII (pp. 9-12) 

International Projectionist 

14 (July, 1939), No. 6 
Supplementary Aids to Servicing Sound Reproducing 

Systems (pp. 7-10, 25-26) 
Panoramic Projection Equipment (pp. 13-14) 

Baird Theatre Television Receiver (p. 21) 
New Academy Theatre Test Reels (pp. 22-24) 


21 (June, 1939), No. 6 

Die Messfilmeinrichtung beim Weltrekordflung mit der 
He 112U der Heinkel-Werke (Installation of Time 
Recorder on Film on the World Record Flight of 
Heinkel-Werke's He 112U) (pp. 141-143) 

Hilfsgerate zur sensitometrischen Ueberwachung der 
Filmverarbeitung (Accessories for Sensitometric 
Supervision of Film Manufacture) (pp. 143-146) 

Optische Systems fur Ton-Abtastung (Optical System 
for Sound Scanning) (pp. 147-150) 

Artgleichung (Comparison Tables) (pp. 150-154) 

Grundsatzliches iiber Belichtungsmesser (Fundamentals 
of Exposure Meters) (pp. 154-155) 

Zur Farbtemperatur/Kerzenstarkcharakteristtk von 
Wolframfadenlampen (Color Temperature Candle 
Power Characteristic of Tungsten Lamps) (pp. 155- 

21 (July, 1939), No. 7 

Tonaufzeichnung in Doppelzackenschrift auf 16-mm- 
Filme (Sound Recording on 16-Mm Films with Double 
Edged Variable Width Track) (pp. 167-172) 
ur Sensitometrie der Umkehrentwicklung von Tonauf- 
zeichnungen (On the Sensitometry of Reversal De- 
velopment of Sound Records) (pp. 172-174) 
ie Konstruktion in der Schmalfilmtechnik (Construc- 
tion in Substandard Film Technique) (pp. 175-178) 
.e neue 8-mm-Kinokamera (A New 8-Mm Motion 
Picture Camera) (pp. 178-181) 


















Schmalfilmaufnahme- und Wiedergabegerate (List of H. FICHTNER 
Substandard Sound Film Cameras and Projections) 
(pp. 191-193) 

Motion Picture Herald 

136 (July 8, 1939), No. 2 

Majors Formally Enter School Film Business with 600 

Motion Picture Herald (Better Theatres Section) 

136 (July 22, 1939), No. 4 
Advancing the Art of Motion Picture Presentation 

(pp. 6-7, 19) 

Auditorium Designs to Serve Both Sound and Decora- 
tion (pp. 8-10) 
The Cost of Projection Light (pp. 17-18) H. D. BEHR 

Philips Technical Review 

4 (June, 1939), No. 6 
Synthetic Sound (pp. 167-173) J. F. SCHOUTEN 

RCA Review 

4 (July, 1939), No. 1 

Application of Motion-Picture Film to Television E. W. ENGSTROM, 
(pp. 48-61) G. L. BEERS, AND 

A Push-Pull Ultra-High-Frequency Beam Tetrode A. K. WING 

(pp. 62-72) 
An Iconoscope Pre-Amplifier (pp. 89-107) A. A. BARCO 



Officers and Committees in Charge 

E. A. WILLIFORD, President 

S. K. WOLF, Past-President 

W. C. KUNZMANN, Convention V ice-President 

J. I. CRABTREE, Editorial Vice-President 

D. E. HYNDMAN, Chairman, Atlantic Coast Section 
J. HABER, Chairman, Publicity Committee 

S. HARRIS, Chairman, Papers Committee 

H. GRIFFIN, Chairman, Convention Projection 

E. R. GEIB, Chairman, Membership Committee 

Reception and Local Arrangements 

D. E. HYNDMAN, Chairman 







Registration and Information 

W. C. KUNZMANN, Chairman 


Hotel and Transportation 

J. FRANK, JR., Chairman 

C. Ross P. D. RIES 



J. HABER, Chairman 


P. A. McGuiRE 

P. A. McGuiRE 


462 1939 FALL CONVENTION fj. s. M. p. E. 

Convention Projection 

H. GRIFFIN, Chairman 






Officers and members of Projectionists Local 306, IATSE 

Banquet and Dance 

A. N. GOLDSMITH, Chairman 





Ladies 1 Reception Committee 

MRS. O. F. NEU, Hostess 






Headquarters. The headquarters of the Convention will will Hotel Pennsyl- 
vania, where excellent accommodations have been assured, and a reception suite 
will be provided for the Ladies' Committee. 

Reservations. Early in September room reservation cards will be mailed to 
members of the Society. These cards should be returned as promptly as possible 
in order to be assured of satisfactory accommodations. The great influx of visi- 
tors to New York, because of the New York World's Fair, makes it necessary to 
act promptly. 

Hotel Rates. Special per diem rates have been guaranteed by the Hotel Penn- 
sylvania to SMPE delegates and their guests. These rates, European plan, will 
be as follows: 

Room for one person $ 3.50 to $ 8.00 

Room for two persons, double bed $ 5.00 to $ 8.00 

Room for two persons, twin beds $ 6 . 00 to $10 . 00 

Parlor suites: living room, bedroom, $12.00, $14.00, and 
and bath for one or two persons $15.00 

Parking. Parking accommodations will be available to those who motor to 
the Convention at the Hotel Fire Proof Garage, at the rate of $1.25 for 24 
hours, and $1.00 for 12 hours, including pick-up and delivery at the door of the 

Oct., 1939] 1939 FALL CONVENTION 463 

Registration. The registration desk will be located at the entrance of the 
Banquet Room on the ballroom floor where the technical sessions will be held. 
All members and guests attending the Convention are expected to register and 
receive their badges and identification cards required for admission to all the 
sessions of the Convention, as well as to the Capitol Theater, Paramount Theater, 
Radio City Music Hall, Roxy Theater, and Warner's Strand Theater. 

Luncheon and Banquet 

The usual informal get-together luncheon will be held in the Grand Ballroom of 
the Hotel on Monday, October 16th. The Semi-Annual Banquet and Dance will be 
I held in the Grand Ballroom of the Hotel Pennsylvania on Wednesday, October 
18th, at 8:30 P. M. At the banquet the annual presentation of the SMPE Prog- 
ress Medal and the Journal Award will be made, and the officers-elect for 1940 
will be introduced. 


Motion Pictures. At the time of registering, passes will be issued to the dele- 
i gates of the Convention admitting them to the motion picture theaters named 

Golf. Golfing privileges at country clubs in the New York area may be ar- 
ranged at the Convention headquarters. In the Lobby of the Hotel Pennsylvania 
will be a General Information Desk where information may be obtained regard- 
ing transportation to various points of interest. 

Miscellaneous. Many entertainment attractions are available in New York to 
the out-of-town visitor, information concerning which may be obtained at the 
General Information Desk in the Lobby of the Hotel. 

Ladies' Program 

A specially attractive program for the ladies attending the Convention is being 
arranged by Mrs. O. F. Neu, Hostess, and the Ladies' Committee. A suite will 
be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. 

New York World's Fair 

\ Members are urged to take advantage of the opportunity of combining the 
(Society's Convention and the New York World's Fair on a single trip. Informa- 
|tion on special round-trip railroad rates may be obtained at local railroad ticket 
offices. Trains directly to the Fair may be taken from the Pennsylvania Station, 
opposite the Hotel: time, 10 minutes; fare, 10^. Among the exhibits at the 
Fair are a great many technical features of interest to motion picture engineers. 


Points of Interest 

Among the points of interest to the general sightseer in New York may be 
listed the following: 

Museum of Modern Art Film Library. As part of the summer exhibition at 
the Museum of Modern Art, "Art in Our Time," the Museum Film Library is 
conducting "A cycle of Seventy Films" from 1895 to 1935. The Museum is 
open to the public daily from 10 A. M. to 6 p. M. The film showings are daily at 
4 p. M. Admission to the Museum at a nominal charge; to the film showings, 
no charge. 

Metropolitan Museum of Art. Fifth Ave. at 82nd St.; open 10 A. M. to 5 p. M. 
One of the finest museums in the world, embracing practically all the arts. 

New York Museum of Science and Industry. RCA Building, Rockefeller Cen- 
ter; 10 A. M. to 5 P. M. Exhibits illustrate the development of basic industries, 
arranged in divisions under the headings food, industries, clothing, transportation, 
communications, etc. 

Hayden Planetarium. Central Park West at 77th St. Performances at 1 1 A. M., 
2 p. M., 3 P. M., 4 P. M., 8 P. M., and 9 P. M. Each presentation lasts about 
45 minutes and is accompanied by a lecture on astronomy. 

Rockefeller Center. 49th to 51st Sts., between 5th and 6th Aves. A group of 
buildings including Radio City Music Hall, the Center Theater, the RCA Build- j 
ing, and the headquarters of the National Broadcasting Company, in addition toi 
other interesting general and architectural features. 

Empire State Building. The tallest building in the world, 102 stories or 1250 j 
feet high. Fifth Ave. at 34th St. A visit to the tower at the top of the building \ 
affords a magnificent view of the entire metropolitan area. 

Greenwich Village. New York's Bohemia; a study in contrasts. Here are| 
located artists and artisans, some of the finest homes and apartments, and some 
of the poorest tenements. 

Foreign Districts. Certain sections of the city are inhabited by large groups of 
foreign-born peoples. There is the Spanish section, north of Central Park; the 
Italian district near Greenwich Village; Harlem, practically a city in itself, num- j 
bering 300,000 negroes; Chinatown, in downtown Manhattan; the Ghetto, the 1 
Jewish district; and several other such sections. 

Miscellaneous. Many other points of interest might be cited, but space permits 
only mentioning their names. Directions for visiting these places may be ob-| 
tained at the Convention registration desk: Pennsylvania Station, Madison j 
Square, Union Square, City Hall, Aquarium and Bowling Green, Battery Park,; 
Washington Square, Riverside Drive, Park Avenue, Fifth Avenue shopping dis-i 
trict, Grand Central Station, Bronx Zoo, St. Patrick's Cathedral, St. Paul's! 
Chapel, Cathedral of St. John the Divine, Trinity Church, Little Church Around! 
the Corner, Wall St. and the financial district, Museum of Natural History : 
Columbia University, New York University, George Washington Bridge, Brook- 
lyn Bridge, Triborough Bridge, Statue of Liberty, American Museum of Natural 
History, Central Park, Metropolitan Museum of Art, and Holland Tunnel. 



OCTOBER 16-19, 1939 

The Papers Committee submits for the consideration of the membership the follow- 
ing abstracts of papers to be presented at the Fall Convention. It is hoped that the 
{publication of these abstracts will encourage attendance at the meeting and facilitate 
\ discussion. The papers presented at Conventions constitute the bulk of the material 
published in the Journal. The abstracts may therefore be used as convenient refer- 
i ence until the papers are published. 

J. I. CRABTREE, Editorial Vice-P resident 

S. HARRIS, Chairman, Papers Committee 

L. A. AICHOLTZ, Chairman, West Coast Papers Committee 







Photographic Duping of Variable-Area Sound; F. W. Roberts and E. Taenzer, 

Ace Film Laboratories, Brooklyn, N. Y. 

In release print laboratories it is necessary to have some method of quickly 
making duplicate sound negatives which are used to replace damaged original 
negative sections. New negative may, of course, be re-recorded from a release 
print, but inasmuch as recording equipment is not always available, a suitable 
photographic process had to be developed. For this process, the following criteria 
were set up : 

The quality of sound from the dupe negative should be high, so that a trained 
observer would have difficulty in telling where a dupe had been inserted. All 
developing should be done in the regular release print positive bath at standard 
developing time. Inasmuch as this bath is in constant use, no special machine 
need be started to develop a dupe. The dupe negative must have the same 
optimum print density as the original negative, and the same fog value in order 
that the inserted dupe may be printed on the same printer light as the original. 

The method that was developed operates as follows : Master positives of every 
reel of a release and the accompanying cross-modulation tests are first printed on 
high-contrast title stock. A density of 2.20 is used, a family of cross-modulation 
curves having indicated this value as best. The reels of master positive are 
stored, but the cross-modulation test positives are detached and printed on 
regular positive stock to make dupe negative cross-modulation tests. The test 



from reel 1A is printed at three negative densities, and the tests from the remain- 
ing reels are printed to a density of about 1.80. Cross-modulation prints at several 
densities are then made from each of the dupe negative cross-modulation tests, 
and from these prints the optimum print density for each dupe negative test is 
determined. Reel 1A gives a three-point slope curve of negative density vs. print 
optimum density. 

The print optima of the dupe negative tests are now compared with the print 
optima of the original negative tests (these data being on file) . If the dupe values 
are different from those of the original, the slope curve of 1A is used to find the 
negative density that will yield a print density the same as that of the original, 
and these values of corrected negative densities are kept on file for use when it 
becomes necessary to make a dupe from the stored master positive. 

The paper includes a complete cross-modulation treatment of the subject and a 

A Sound-Track Center-Line Measuring Device; F. W. Roberts and H. R. 
Cook, Jr., Ace Film Laboratories, Brooklyn, N. Y. 

The types of instruments now in use for measuring the position of sound- 
tracks on film are not completely suited to the use of a release print laboratory. 
Microscopes using micrometer stages or oculars are slow in operation because they 
require mental arithmetic to arrive at the distance from the film edge to the sound- 
track center-line. Projection types are slow in threading; and require a darkened 
room. The release print laboratory requires a small quick-threading device which 
gives a direct reading of sound-track position. 

A device that fulfills these requirements has been built, and consists of a 
curved film-gate in which the film is held against a guiding edge by means of a 
spring parallel. This gate is mounted in F-slides which permit motion in a direc- 
tion perpendicular to the length of the film. Motion is provided by a hand-lever- 
operated cam, and the position of the gate is measured by a one-ten-thousandtt 
dial indicator. 

The gate has in it a hole directly under the sound-track, and beneath is mounted 
a small incandescent lamp. Directly above the gate is an optical system consist 
ing of a standard 32-mm microscope objective and a 10-power Huygen's eyepiece 
The normal cross-hairs of the eyepiece have been replaced with a parallel haii 
device consisting of two very fine hairs whose mounts slide hi F-ways perpen 
dicular to the direction of the film. Both hairs operate together and are operatec 
by a common cam and lever which cause them to move ; and as they separate 01 ; 
close, they always remain parallel to each other and equidistant from the optica 
center of the instrument. 

The operation is as follows : With film hi the gate the operator places a hanc 
on each of the two levers, which are moved simultaneously until the two cross 
hairs are directly over the bias lines or over corresponding peaks of modulatec 
variable-area track. The one-ten-thousandth dial indicator then indicates thcj 
track center-line position to the nearest ten-thousandth of an inch. With the; 
instrument, a film may be inserted and a reading taken in ten seconds. 

Volume Distortion; S. L. Reiches, Cleveland, Ohio. 

The contention that a linear recording and reproducing system represents th< 


ideal, and that sound handled by such a system will be exactly represented, is not 
borne out by experience. Systems have been built which meet this requirement 
within limits that are not detectable by the ear and yet these systems do not 
reproduce sound as it actually is produced. In many cases a definite non-linear 
response curve is provided to compensate for some factor that is not covered by 
the above contention. It is the author's thesis that this discrepancy is due to the 
ear sensitivity to frequencies as a function of loudness. 

Using the ear sensitivity curves presented by Fletcher and Munson of the Bell 
Telephone Laboratories (which have been verified by other observers) it is shown 
how the ear introduces frequency distortion to a linear system when the sound is 
reproduced at a level other than the level at which it is produced. It is shown 
how a sound reproduced above the incident sound level introduces excessive low 
frequencies. The case for a sound reproduced at a lower level is also examined 
and the conclusion is drawn that this case accentuates the high frequencies. 

It is further shown that the possibility of correcting for the limited volume range 
of all sound systems may lie in the type of amplifier response curve. 

A description is given of three methods used to achieve the desired amplifier 
characteristics: (1) a mechanical method, (2) a linear-non-linear system, and (5) 
a selective by-pass system. Circuits are given and the important operating points 
of each are discussed. The objections to each system are also given. 

Further, a brief summary, with diagrams, describes the various set-ups used 
to record with these amplifiers. This covers work for radio, disk record, and 

Television Control Equipment for Film Transmission; R. L. Campbell, Al- 
len B. DuMont Laboratories, Passaic, N. J. 

A television film chain with particular reference to amplifier, sweep, and power 
circuits in the film pick-up unit is described. 

Many improvements in television circuits have been made possible by recent 
advances in circuits and circuit components in radio and allied electronic fields. 
Application of some of the newer ideas to motion picture film pick-up equipment 
has resulted in improved performance and simplicity of operation. 

Circuit arrangements which permit flexibility in transmission standards are 
considered and their application discussed. Also the anticipation of possible 
future improvements in picture quality is indicated in some circuit capabilities. 

Simplification of controls from the television projectionist's standpoint is dis- 

The Production of a Three-Dimensional Motion Picture; J. A. Norling, 
Loucks and Norling, New York, N. Y. 

Some problems involved in the production of satisfactory three-dimensional 
motion pictures have not received much mention in the literature dealing with 
stereoscopy. Their practical solution has contributed marked improvements to 
the three-dimensional picture of today. 

The fundamental problem in projecting three-dimensional pictures is that of 
providing a "right-eye" picture that will reach only the right eye and be prevented 
from reaching the left eye, and to do the same for the "left-eye" picture. To 
attain this result two methods have been employed with success, namely: the 


"anaglyph" in which substantially complementary colors are employed in the 
viewing devices, and polarized light. 

The screen surface upon which three-dimensional pictures are projected by 
polarization methods is of extreme importance. The selection of the proper type 
of screen raises real problems but these also have been overcome in a practical 

Considerations Relating to Warbled Frequency Films; E. S. Seeley, Altec 
Service Corp., New York, N. Y. 

Some warbled frequency films, intended as signal sources for acoustical response 
measurements, appear to have been made and used without full realization of the 
true nature of the warbled signal and the manner in which such a signal is affected 
by a non-linear transmission system. It is pointed out that the warbled signal 
is a frequency-modulated signal; hence the signal may be represented by a car- 
rier frequency and a series of side-frequencies, all of which are steady and discrete. 
It is pointed out, and substantiated experimentally, that the signal must be re- 
garded in this light when considering the effect on it of a non-linear transmission 
system. The frequency structure of one "warble film" in use is calculated and 
shown graphically. Fundamental requirements for a suitable warbled frequency 
film having sinusoidal modulation are discussed and values for modulation rate and 
for modulation depth are recommended. The side-frequency array provided by 
the recommended modulation constants is shown in graph form. Expressions are 
derived giving the frequency relationship and relative amplitudes of the side- 
frequencies resulting from the non-sinusoidal frequency modulation which con- 
tains two components of modulation rate, one component having an associated 
phase constant. The side-frequency structures corresponding to some assumed 
combinations of two rates are calculated and illustrated. Certain assumptions 
are made for distortion or departure from sinusoid of a modulating frequency and 
the effects on the side-frequency structure are shown. From the latter calculation 
recommendations are derived for tolerances of departure from sinusoidal modu- 
lation for a warbled frequency film. 

A Transmission System of Narrow Band- Width for Animated Line Images; 
A. M. Skellett, Bell Telephone Laboratories, New York, N. Y. 

A new method of transmission and reproduction of line images, e. g., drawings, 
is described which utilizes a cathode-ray tube for reproduction, the spot of which 
is made to trace out the lines of the image twenty or more times a second. The 
steps of the complete process are: (1) the transcription of the line image into two 
tracks similar to sound-tracks on motion picture film; (2) the production from 
these tracks of two varying potentials by means of photoelectric pick-up devices; 
(5) the transmission of these potentials ; and (4) their application to the cathode- 
ray deflector plates to effect reproduction. Satisfactory transmission of fairly 
complex images, e. g., animated cartoons, could be effected within a total band 
width of 10,000 cycles. 

Science and the Motion Picture; H. Roger, Rolab Photo-Service Laboratories, 
Sandy Hook, Conn. 


The motion picture is a product of science. There is ample historical material 
available for those who wish to convince themselves of this fact, but a brief re- 
view is given of the work of Muybridge and Marey in order to clarify the cause 
of their inventions. The ensuing discussion centers around the question, "Has 
science maintained its interest in the motion picture and has it utilized its ad- 
vantages to its full extent?" 

In this paper the word "science" is taken broadly and includes research, dis- 
semination of knowledge, and industrial application. Motion picture's applica- 
tion to science is divided into two distinct categories and are discussed in detail : 

(7) The motion picture as an aid to scientific research; 

(2) The motion picture as a medium for the dissemination of knowledge. 

The paper concludes with descriptions and demonstrations of interesting ma- 
terial from the files of the Rolab Photo-Science Laboratories. Also an inside 
view is given of production activities of an unusual character. 

The Problem of Distortion in the Human Ear; S. S. Stevens, Harvard Uni- 
versity, Cambridge, Mass. 

The amount of distortion produced by the ear upon a simple sound-wave has 
been measured by analyzing the electrical output of the ears of animals and by in- 
direct experiments with human ears. The amount of distortion in a sound-wave 
which the human ear is just able to detect has also been determined, and it is found 
that the threshold of audible distortion is intimately related to the amount of 
distortion occurring in the ear itself. Hence the transmission characteristics of 
the ear determine the tolerances for distortion in sound reproduction. 

Report of the Standards Committee; E. K. Carver, Chairman. 

Proposals have been received from the ISA Secretariat for International 
Standardization of raw-film cores; 16-mm sound-film; projection reels; projec- 
tion reel boxes; 8-mm film dimensions; and definition and marking of safety 

Most of these proposals differ from the SMPE standards only in tolerance. 
Some of the tolerances appear to be unimportant and some important. The 
European practice for projection reels differs so widely from the American prac- 
tice that it is deemed impossible to come to an international agreement. Stand- 
ardization of 16-mm projection reel boxes appears to be outside the range of use- 
ful standardization. 

The international standard definition of safety film has been cleared up in all 
points except the question of nitrogen content. 

The question of sound-track dimensions for 35-mm and 16-mm film was clari- 
fied, to a considerable extent, at the Hollywood meeting of the Committee but 
no definite conclusions have yet been reached. 

No satisfactory standard for 16-mm sound-film sprockets has yet been at- 

The publication of the Academy standard 2000-f t release print has been delayed 
pending further questions by the Academy. 

Some Industrial Applications of Current 16-Mm Sound Motion Picture Equip- 
ment; W. H. Offenhauser, Jr., and F. H. Hargrove, The Berndt-Maurer Corp., 
New York, N. Y. 


Sixteen-mm sound motion pictures are potentially one of the most effective 
means through which industry can develop a broad, cost-cutting communication 
system within the organization itself. 

Many latent applications for internal films exist ; the cases in business where the 
improved transfer of ideas afforded by films can be most profitable are almost un- 
limited. Several specific instances are cited. 

Sixteen-mm equipment is simple, easy to operate, reliable, and economical.. 
With it, a member of the industrial organization who knows his company's prod- 
ucts, policies, and structure can readily produce films that are, in every respect, 
profitable internal communications media. 

Future Development in the Field of the Projectionist; A. N. GOLDSMITH, New 
York, N. Y. 

The highly diversified activities required for the production of a motion picture 
find their effective culmination in the work of the theater projectionist. The un- 
usually concentrated value embodied in the reels of film corresponding to a feature 
picture can be brought to the theater audience and made the basis for commercial 
returns only through the activities of the projectionist. 

Nevertheless the public is little aware of what goes on in the projection room. 

The projectionist is in part compensated by the likely stability of his activities. 
His present position in the theater is important. Future developments in the mo- 
tion picture field, such as three-dimensional sound, wider use of color, and the 
like, will make his work even more important. The possible inclusion of television 
projection in theater programs will require his mastery of the new field which is 
sufficiently similar to his present activities in its broad outline to enable its han- 
dling by the theater projectionist. 

The Projectionist's Part in Maintenance and Servicing; J. R. Prater, Con- 1 
g ress Theater, Palouse, Wash. 

It is the duty of the projectionist to see that all projection equipment is kept J 
in condition to give excellent service dependably and efficiently. It is impossible 
to accomplish these results by depending upon memory alone. The projectionist 
must establish and keep written records of all necessary maintenance data. He 
must follow a written schedule in making inspections and in doing maintenance 
work. He must establish a reliable system for checking and ordering supplies 
and spare parts at regular intervals. 

The projectionist should do as much of actual service work as his knowledge, 
ability, tools, and available test equipment will permit. At least nine-tenths of 
trouble shooting should be done before any trouble exists. He should obtain de- 
tailed drawings of internal and installation wiring of ail electrical equipment, be- 
sides identifying the points at which tests may be made. He should prepare a 
written outline of all tests that could be made if various troubles existed. Then 
he should actually make all possible tests in advance, wherever possible, without 
causing damage, by deliberately creating the trouble and then correcting it. He 
should immediately record the exact results of each test in the written outline. 
In this way, simple tests may serve as well as or better than elaborate ones. 

The professional service engineer with special test equipment is a necessity to 
the finer and more difficult parts of modern servicing, but the projectionist who! 


makes the best of what resources he has can also do a very valuable part of the 

Suggestions for Encouraging Study by Projectionists; F. H. Richardson, 
Motion Picture Herald, New York, N. Y. 

This paper stresses the great importance of expert work in theater projection 
rooms and points out that pride in performance is essential to high excellence. If 
the status of projection were elevated to a higher plane there would be as a result 
improved excellence in results both on screen and through loud speakers. It 
offers a suggestion concerning the contacts of the Society with the projectionists' 

The Production of 16-Mm Sound Pictures for Promoting Safety in the Mineral 
Industries; M. J. Ankeny, Bureau of Mines, U. S. Department of the Interior, 
Pittsburgh, Pa. 

The paper deals chiefly with experience in developing 16-mm direct sound 
recording technic in producing safety educational films. Attention is called to 
the fact that direct 16-mm recording and photography have a great potential use- 
fulness in the field of education, not as a competitor of 35-mm, but as a means of 
extending the use of sound motion pictures into fields that 35-mm is unable to 

Some of the difficulties encountered in underground motion picture photog- 
raphy and how these difficulties were met are described; also the types of film 
used and the various printing methods that have been employed in order to 
arrive at a most satisfactory procedure. 

The method employed in recording sound directly on 16-mm film, in which 
the double-system variable-area is used, is described in some detail. 

Artificial Reverberation for Motion Picture Studios; P. C. Goldmark and 
P. S. Hendricks, Columbia Broadcasting System, New York, N. Y. 

An electrooptical method of producing reverberation synthetically will be de- 
scribed and the latest model of the equipment will be demonstrated. The 
method employed consists basically of recording the original program on the rim 
of a phosphor-coated disk by means of a modulated light-source and then pick- 
ing up the continuously decaying sound images after a predetermined time inter- 
val by means of photocells. 

The exponential decay curve of the phosphorescent substance will produce an 
infinite number of secondary sound impulses to which any desired decay char- 
acteristic can be applied. This reverberation signal is then mixed with the original 
program in the proportion required. 

This new reverberation device has been successfully employed in radio broad- 
casting and can be used in phonograph as well as in motion picture sound re- 
cording, where the scenic effect or script requires a type of sound which, due to 
the deadness of the sound stage, might not readily be available. 

This synthetic reverberation device would replace the use of so-called echo 
chambers, at the same time introducing an appreciable amount of flexibility with- 
out degrading the quality of the original sound. 


Delivering Laboratory Results to Theater Patrons; J. R. Prater, Congress 
Theater, Palouse, Wash. 

A discussion emphasizing the importance of actqally delivering the benefits 
of laboratory research and developments to the theater patrons who furnish the 
financial support for practically the entire motion picture industry. 

Accomplishments in photography, sound recording, projection, and sound 
reproduction are discussed briefly. Examples are given of various ways in which 
theater screen results may suffer regardless of the excellence of films and equip- 

It is pointed out that whatever can be done to increase the projectionist's 
technical knowledge, ability, and pride in good workmanship will ultimately bene- 
fit the entire industry. To this end, it is suggested that if possible, information 
from the JOURNAL of the Society of Motion Picture Engineers be made easily 
available to projectionists. 

A New Non-Intermittent Motion Picture Projector; F. Ehrenhaft and F. G. 
Back, New York, N. Y. 

The authors have designed a projector wherein the optical compensation is 
effected by means of a rotating glass prism. The problem was originally at- 
tacked from the viewpoint that by eliminating the errors inherent in the rotating 
glass prism, a projector could be designed that would be both simple and practi- 
cable. The dimensions of the rotating glass prism and its optical placement result 
from basic optical laws, and the arrangement depends upon the size of the image 
and on the materials. Errors inherent in the rotating glass prism are (1) Non- 
linear displacement on the screen causing a lack of definition: (a) errors of the 
center rays, (b) errors of the corner rays, (c) errors caused by shrinking of the 
film; (2) Chromatical errors; (3) Spherical errors: (a) caused by the size of the 
prism, (&) caused by the deviation of light in glass; (4) Astigmatism caused by 
the movement of the prism; (5) Side images (projection of more than one frame 
on the screen); (6) Limited focus; (7) Defects by reflection. 

Elimination of these errors was achieved by: (1) (a) Limitation of the effec- 
tive rotation angle, (6) use of a curved gate, (c} establishing the tolerable limits of 
film shrinkage; (2) Calculating size and displacement of the colors at the extreme 
position of the prism; (3) (a) Use of special lenses or additional lenses corrected 
for glass instead of for air, (6) compensation by a curved gate; (4) Slip-shaped 
diaphragms ; (o) Use of diverse diaphragms; (6) Use of special lenses or additional 
lenses; (7) Diaphragms for the condenser and screening off the edges of the 
rotating prism. Relation between amount of light on the screen, absence of 
flicker, and arrangement of condenser and lamp-filament. 

These factors will be treated by means of illustrations and diagrams. A work- 
ing model will be shown and test-films projected to illustrate what has been 
accomplished up to now. 

A Flexible Time-Lapse Outfit; W. W. Eaton, Eastman Kodak Company, 
Rochester, N. Y. 

An apparatus is described which has been designed to enable single movie 
frames to be made automatically at intervals conveniently adjustable over a wide 
range. It is known as the Electric Time-Lapse Outfit, and is designed primarily 


for the Cine-Kodak Special. It consists of an electromagnet which mounts on 
the camera and interacts with the one-frame shaft, and suitably housed condenser- 
resistance circuits which supply impulses to the electromagnet and cause pictures 
to be taken. By expanding basic times through an interval multiplier, pictures 
may be made automatically at intervals throughout a range of 1 / i second to 24 
hours. The exposure time is completely independent of the time between pic- 
tures and may be set throughout a range of Vioo second to 6 seconds. In addi- 
tion, where artificial illumination is required, a lamp control is provided which 
automatically turns the lights on and off for each exposure, regardless of the time 
between pictures. The whole outfit operates on self-contained batteries, and is 
entirely portable. 

Automatic Slide Projectors for the New York World's Fair; Fordyce E. Tuttle, 
Development Dept., Eastman Kodak Co., Rochester, N. Y. 

Special slide-changing projectors were designed and built for the Kodachrome 
exhibit in the Eastman building at the New York World's Fair. The individual 
screen images are seventeen feet wide and twenty- two feet high. Eleven machines 
are synchronized so that panoramic scenes one hundred and eighty-seven feet 
long may be shown. Indexing of the slides is controlled by notches in a sound- 
film so that the entire program is automatic. 

The slides in each machine are arranged in two rows, and each machine has two 
gates and two complete optical systems. All the slides in one row are rigidly 
bolted to a ring-gear forty-eight inches in diameter. For each new picture the 
ring-gear is spring-indexed into a new position. While one gear is being moved 
the other is stationary, and the picture being projected is in the stationary row. 
An optical compensator geared to the ring-gear corrects for any inaccuracies in 
indexing, and the image is optically "dowelled" on the screen. The accuracy of 
registration is such that one slide may be substituted for another without move- 
ment on the screen. 

The light-source used is a 2500-watt, high color- temperature tungsten lamp. 
Water-cells and refrigerated air are used to cool the film in the gates. The shutter 
system is located between the lamp and the gate in order to minimize the heat at 
the gate. Shutters in the two beams are interlocked in such a way that while 
they are being moved the light to the screen is constant. The cross-dissolve may 
be rapid or slow depending on the type of transition desired. 

Slide projectors similar in structure are also being used in the Perisphere Build- 
ing. There the slides are projected in rapid enough succession to show motion. 

Motion Picture Theater Auditorium Lighting; B. Schlanger, New York, N. Y 
The various functions of motion picture theater auditorium lighting are dis- 
cussed. Particular analysis is made of the lighting which is used during the period 
in which the motion picture is projected. Past and present lighing practices in 
this respect are explained. The advantages and disadvantages of these practices, 
and a new type of lighting are discussed. It is proposed that the illumination 
levels of the interior surface of the auditorium be at greater levels than have been 
heretofore found to exist. A definite relationship between the screen brightness 
and that of the auditorium surfaces is indicated as desirable. Recent tendencies 
toward higher screen brightnesses have made a very low intensity lighting in the 


auditorium much more undesirable, and therefore have made it more important 
to arrive at a new solution for motion theater auditorium lighting. The realism 
of the projected picture can be considerably heightened by proper surface illumi- 
nation. Controlled reflected light coming from the screen and re-reflected from 
the interior surfaces is discussed as a medium for lighting. 

Lenses for Amateur Motion Picture Equipment (16-Mm and 8-Mm); R. Kings- 
lake, Eastman Kodak Company, Rochester, N. Y. 

In all motion picture photography and projection, lenses of high relative aper- 
ture must be used. However, on account of the small size of the amateur frame, 
the focal length is short, and the linear aperture of the lens is therefore small, 
resulting in considerable depth of field. Thus in cine work, great lens speed is not 
automatically associated with small depth, as is the case in ordinary photography. 

Moreover, as the entire motion picture frame must be seen by the eye at a glance, 
the angular field covered must be much smaller than in still pictures which may 
be examined critically and deliberately. This fact is of the greatest assistance to 
the lens designer because high aperture and field are inevitably somewhat in- 
compatible, and types of lens construction which favor aperture generally cover 
a relatively small field. 

Perspective considerations usually require a projection lens covering only about 
half the angular field covered by the taking lens, which fact enables projection 
lenses of very high relative aperture to be made. Some of the types of construc- 
tion commonly used in amateur cine lenses are described, including an account of 
the Kodak line of 16-mm and 8-mm lenses. 

Tape Splicers for Film Developing Machines; J. G. Capstaff and J. S. Beggs, 
Eastman Kodak Co., Rochester, N. Y. 

The splicers described make a strong, waterproof tape splice for sprocketless 
developing machines, and have proved very successful. The ends of the films to 
be spliced are placed one at a time in a punch and die, where three holes are 
punched on the center line of the films, the ends of which are then trimmed. The 
two holes away from the ends are placed over two pairs of pegs, which are in a 
straight line and spaced so that the abutting ends of the film are separated by 
Vie of an inch. A piece of 1-inch wide waterproof adhesive tape, previously 
placed adhesive-side up and symmetrically under this space, is now wrapped 
around the film. The tape sticks to itself through the two holes near the ends of 
the film, thus preventing the tape from loosening when the emulsion swells. The 
splice is quite thin, and there is nothing about it to catch in the machine or blow- 
off. It is also very dependable and will not mar film in taking up or make the roll 
out of round. 

An Investigation of the Ground-Noise of Photographic Sound Prints; O. Sand- 
vik and W. K. Grhnwood, Eastman Kodak Co., Rochester, N. Y. 

This paper deals with the effect of the negative sound-track on the ground-noise 
of the print. Data are presented showing the influence of negative density and 
negative gamma on print ground-noise for fine, medium, and coarse-grain nega- 
tive emulsions. 


The Backward Perspective a Resume of Three Years of Progress in the Film 
Library of the Museum of Modern Art; Douglas L. Baxter, Museum of Modern 
Art Film Library, New York, N. Y. 

In 1936 a paper was read before the Society outlining the organization and aims 
I of the Museum of Modern Art Film Library, then in its first year of existence. 

The present paper deals with the activities of the Film Library during the 
intervening three years, giving a brief resum of its growth in the local, national, 
and international fields, and tracing its development from the position of an un- 
recognized institution with rather limited headquarters to one where the opening 
of its permanent headquarters in its own $2,000,000 building was considered of 
sufficient national importance for the President of the United States to make it 
the occasion of a nation-wide broadcast. 

In international circles the importance of the Film Library has won such recogni- 
tion that in response to its invitations representatives of twelve foreign nations as 
well as representatives of the Library of Congress and the Division of Cultural 
JRelations of the State Department attended the first annual Congress of the 
ilnternational Federation of Film Archives held in New York in July, 1939. 

The present activities and future plans of the Film Library are explained. The 
paper is amplified by the showing of The Movies March On, a recent issue of the 
"March of Time," devoted to the work of the Museum of Modern Art Film Li- 

Statement of the Ownership, Management, Circulation, Etc., Required by the 

Acts of Congress of August 24, 1912, and March 3, 1933, of Journal of the Society 

of Motion Picture Engineers, published monthly at Easton, Pa., for October 1 


State of New York \ 

County of New York / * 

Before me, a Notary Public in and for the State and County aforesaid, person- 
ally appeared Sylvan Harris, who, having been duly sworn according to law, 
deposes and says that he is the Editor of the Journal of the Society of Motion 
Picture Engineers and that the following is, to the best of his knowledge and 
belief, a true statement of the ownership, management (and if a daily paper, 
the circulation), etc., of the aforesaid publication for the date shown in the above 
caption, required by the Act of August 24, 1912, as amended by the Act of 
March 3, 1933, embodied hi section 537, Postal Laws and Regulations, printed 
on the reverse of this form, to wit : 

1. That the names and addresses of the publisher, editor, managing editor, 
and business managers are: 

Name of Post Office Address 

Publisher, Society of Motion Picture Engineers, Hotel Pennsylvania, New York 


Editor, Sylvan Harris, Hotel Pennsylvania, New York, N. Y. 
Managing Editor, Sylvan Harris, Hotel Pennsylvania, New York, N. Y. 
Business Manager, Sylvan Harris, Hotel Pennsylvania, New York, N. Y. 

2. That the owner is: (If owned by a corporation, its name and address 
must be stated and also immediately thereunder the names and addresses of 
stockholders owning or holding one per cent or more of total amount of stock. 
If not owned by a corporation, the names and addresses of the individual owners 
must be given. If owned by a firm, company, or other unincorporated concern, 
its name and address, as well as those of each individual member, must be given). 
Society of Motion Picture Engineers, Hotel Pennsylvania, New York, N. Y 

E. A. Williford President, 30 East 42nd St., New York, N. Y. 
J. Frank, Jr., Secretary, 356 W. 44th St., New York, N. Y. 
L. W. Davee, Treasurer, 153 Westervelt Ave., Tenafly, N. J. 

3. That the known bondholders, mortgagees, and other security holders ; 
owning or holding 1 per cent or more of total amount of bonds, mortgages, or 
other securities are: (If there are none, so state). 


4. That the two paragraphs next above, giving the names of the owners, 
stockholders, and security holders, if any, contain not only the list of stockholders 
and security holders as they appear upon the books of the company but also, 
in cases where the stockholder or security holder appears upon the books of the 
company as trustee or hi any other fiduciary relation, the name of the person or 
corporation for whom such trustee is acting, is given; also that the said two 
paragraphs contain statements embracing affiant's full knowledge and belief 
as to the circumstances and conditions under which stockholders and security 
holders who do not appear upon the books of the company as trustees, hold stock 
and securities hi a capacity other than that of a bona fide owner; and this affiant 
has no reason to believe that any other person, association, or corporation has 
any interest direct or indirect hi the said stock, bonds, or other securities thai) 
as so stated by him. 

5. That the average number of copies of each issue of this publication sold 
or distributed, through the mails or otherwise, to paid subscribers during the 
six months preceding the date shown above is: (This information is required 
from daily publications only). 

SYLVAN HARRIS, Editor, Business-Manager. 

Sworn to and subscribed before me this 19th day of September, 1939. 

(Seal) Wm. J. Miller. 

Notary Public, Clerk's No. 180, Ne* 
York County. Reg. No. OM 104 
(My commission expires March 30, 1940) 




Volume XXXIII November, 1939 



A Direct Positive System of Sound Recording 


The Polyrhetor a 150-Channel Film Reproducer 

Effect of Orientation of the Scanning Image on the Quality of 

Sound Reproduced from Variable- Width Records 

D. FOSTER 502 
The Chemical Analysis of Hydroquinone, Metol, and Bromide 

in a Photographic Developer H. L. BAUMBACH 517 

New Frontiers for the Documentary Film A. A. MERCEY 525 

Safekeeping the Picture Industry K. W. KEENE 533 

New Motion Picture Apparatus 

A New Magnetic Recorder and Its Adaptations 

S. J. BEGUN 538 
Modern Instantaneous Recording and Its Reproduction. . . . 

A Newly Designed Sound Motion Picture Reproducing 

Equipment J. S. PESCE 551 

A High-Intensity Arc for 16-Mm Projection H. H. STRONG 569 

New 16-Mm Recording Equipment D. CANADY 571 

Notes on French 16-Mm Equipment D. CANADY 573 

MGM Portable Dolly Channel C. S. PRATT 578 

Simplifying and Controlling Film Travel through a Develop- 
ing Machine J. F. VAN LEUVEN 583 

Current Literature 586 

1939 Fall Convention at New York, October 16th-19th: 

Highlights of the Convention 588 

Program 594 

Abstracts of Convention Papers 597 

Society Announcements 600 





Board of Editors 
J. I. CRABTREE, Chairman 




Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscription or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton. Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

West-Coast Office, Suite 226, Equitable Bldg., Hollywood, Calif. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1939. by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing hi this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers hi question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is not 
responsible for statements made by authors. 


** President: E. A. WILLIFORD, 30 East 42nd St., New York, N. Y. 
** Past-President: S. K. WOLF, RKO Building, New York, N. Y. 
** Executive Vice-President: N. LEVINSON, Burbank, Calif. 

* Engineering Vice-President: L. A. JONES, Kodak Park, Rochester, N. Y. 
** Editorial Vice-President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 

* Financial Vice-President: A. S. DICKINSON, 28 W. 44th St., New York, N. Y. 
** Convention Vice-President : W. C. KUNZMANN, Box 6087. Cleveland, Ohio. 

* Secretary: J. FRANK, JR., 356 W. 44th St., New York, N. Y. 

* Treasurer: L. W. DAVEE, 153 Westervelt Ave., Tenafly, N. J. 

** M. C. BATSEL, Front and Market Sts., Camden, N. J. 

* R. E. FARNHAM, Nela Park, Cleveland, Ohio. 

* H. GRIFFIN, 90 Gold St., New York, N. Y. 

* D. E. HYNDMAN, 350 Madison Ave., New York, N. Y. 

* L. L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* A. C. HARDY, Massachusetts Institute of Technology. Cambridge, Mass. 

* S. A. LUKES, 6427 Sheridan Rd., Chicago, 111. 

** H. G. TASKER, 14065 Valley Vista Blvd., Van Nuys, Calif. 

* Term expires December 31, 1939. 
** Term expires December 31, 1940. 



Summary. One of the fundamental advantages of the variable-area system is that 
the original negative may be reproduced directly without appreciable wave-shape dis- 
tortion. One printer operation can be eliminated and the film noise considerably re- 
duced by originally recording a track in which the transparent area diminishes as the 
recorded level is decreased. A direct positive recording system has been so designed 
that the recording and noise-reduction light-beams are separated in the direction of the 
film motion. This allows the noise-reduction system completely to anticipate a coming 
signal, thereby eliminating clipping. When the sound volume is less than is required 
to fill the track, the noise from the direct recording is reduced further by additionally 
exposing those portions of the track not needed for the sound-waves. A single shutter 
vane automatically controls both the transparent area and the additionally exposed 

A model of the direct positive optical system has been on test in Hollywood for sev- 
eral months. The exposure has proved to be less critical than on the negative-print 
process. Densities from 0.80 to 1.60 produced only minor quality differences. A 
good print of a negative compares favorably in quality with a direct positive record, but 
because of the variations always present in the printing operation, it is believed that a 
higher average of quality can be maintained from the direct record. 

One of the fundamental advantages of the variable-area system 
is that the original negative may be reproduced directly without 
appreciable wave-shape distortion. Although this advantage has 
been recognized from the start, no use could be made of it as long as 
all or part of the final release film was printed from the original re- 
corded negative. 

The present practice, which has been almost universally followed 
by the motion picture industry, is to record a negative, print this to 
a master positive, re-record from the positive to a final negative, and 
print to a release positive. This involves two printer operations with 
the attendant problems of printer contact, slippage and noise. One 
of the printer operations can be eliminated and the film noise con- 
siderably reduced by originally recording a track in which the trans- 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
14, 1939. 

** RCA Manufacturing Co., Camden, N. J., and Hollywood, Calif. 



G. L. DIMMICK AND A. C. BLANEY [J. s. M. p. E. 

parent area diminishes as the recorded level is decreased. The 
final negative can then be re-recorded from this direct positive. 

The exposed areas of a direct recording are opaque and therefore 
relatively free from noise, whereas the exposed areas of a print have a 
large number of minute transparent or partially transparent holes 
which result from opaque particles and abrasions nearly always pres- 
ent in the clear areas of a developed negative. When the sound 

FIG. 1. The sound-waves and corresponding 
noise-reduction waves are recorded simultane- 
ously but may be separated any distance in the 
direction of film motion. 

volume is less than is required to fill the track, the noise from the 
direct recording is decreased still further by additionally exposing 
those portions of the track not needed for the recorded waves. The 
width of the additionally exposed portion is automatically controlled 
by the same shutter and noise-reduction amplifier which control the 
transparent area of the track. The density of the recorded waves is 
usually held to about 1.4 in order to prevent excessive spreading of 
the image and the loss of high frequencies. The additionally ex- 
posed areas are given a density of about 2.2. 

Nov., 1939] 



The width of a standard variable-area sound-track is 76 mils and 
the reproducing slit length is 84 mils. The present practice in print- 
ing is to blacken a strip on both sides of the sound-track so that the 
beam will not hang over into clear film. In making an original re- 
cording for direct reproduction, the recording light beam must be 
longer than 84 mils for the same reason. A total track width of 100 
mils provides enough tolerance to take care of track misplacement 
and film weave. It is desirable, of course, to expose the portions out- 
side the normal track width to a high density. 

With the exception of the class B system, all the noise-reduction 
systems in commercial use at the present time are limited in their 
effectiveness by virtue of the fact that both the sound waves and the 
noise-reduction waves are recorded at the same instant and at the 
same point along the film. This makes it necessary to clip the first 
few waves of a rapidly increasing train of waves such as occurs at the 

FIG. 2. Enlargement of sound-track. 

start of a word or a musical sound. The reduction of noise from the 
print is accomplished by subtracting from either the intensity or the 
length of the recording light-beam. Obviously, the subtracting 
process can be applied only at the light-beam. Schemes have been 
proposed for causing an acoustical or electrical delay of the sound- 
waves so that the noise-reduction system might anticipate the coming 
signal, but so far, practical difficulties have prevented their use. 

The direct positive recording system offers a simple and effective 
method of obtaining anticipation and complete freedom from clip- 
ping. In this system, noise-reduction is accomplished by auxiliary 
light-beams which darken those portions of the track not needed for 
modulation. Since this is an additive process, the noise-reduction 
light-beams may be separated any desired distance from the recording 
beam in the direction of the film motion. The sound-waves and the 
corresponding noise-reduction waves are recorded simultaneously in 
time, but anticipation in space results from the separation. Fig. 1 
shows the order in which the two waves are recorded. The dark area 


G. L. DIMMICK AND A. C. BLANEY [j. s. M. P. E. 

represents the latent image and the clear area represents unexposed 
film. The sound-track enlargement in Fig. 2 shows the action of the 
noise-reduction system when a 1000-cycle signal is suddenly applied 
and indicates the point at which the shutter vane started to move. 
The displacement of the vane was completed at the point where the 
recording of the sine wave began, so that no clipping occurred. The 
sudden application of a full amplitude signal is a very severe test of 


FIG. 3 (Top). Layout of recording optics. 
FIG. 4 (Bottom). Vertical section through axis of galvanometer mirror and 

objective lenses. 

clipping, and a system designed to meet this condition should be 
capable of handling all sounds occurring in nature. 

Fig. 3 shows a layout of the recording optics and Fig. 4 is a vertical 
section taken through the axis containing the galvanometer mirror 
and objective lenses. In addition to the usual recording slit, a 
noise-reduction slit is placed behind the condenser and spaced 200 
mils below the optical axis. This slit is illuminated constantly along 
its full length and a shutter with triangular vane is placed behind it to 
limit the length of the light-beams reaching the film. The noise- 
reduction slit is made wider than the modulation slit in order to ob- 
tain the additional exposure needed for the double density feature. 

Nov., 1939] 



The greater width of this slit is not detrimental since it is required only 
to record low frequencies proportional to the envelope of the sound- 
waves. Two objective lenses are rigidly fastened in a single mounting 
so that they may be moved together and are simultaneously focused 
upon the film. A narrow prism at the noise-reduction slit causes the 
light from this slit to be deviated from its normal path and to fall 
upon the lower objective. An achromatic prism in front of the ob- 
jective again changes the angle of the light so that it passes along the 





FIG. 5. Shapes and relative positions of recording light-beam, shutter vane, and 
slits for class A and class B push-pull systems. 

axis of this lens and to an image of the slit upon the film. Light from 
the modulation slit converges to an image of the galvanometer mirror 
upon the upper objective and then to an image of this slit upon the 
film. A narrow opaque mask placed about midway between the slit 
and the objective lens serves to reduce stray light by making only one 
slit visible from each objective. An ultraviolet filter placed in the 
path of both slits restricts the exposing energy to the 3560 A band. 
The distance between the two slit images at the film is 180 mils 
in the present design. This results in an anticipation of 10 milli- 
seconds. Longer or shorter values of anticipation may be obtained 
by changing the separation distance. 


G. L. DIMMICK AND A. C. BLANEY tf. s. M. P. E. 

Fig. 5 shows the shapes and relative positions of the recording light- 
beam, shutter vane, and slits for both a class A and a class B push- 
pull system. 1 A short section of sound-track is also pictured below 
each diagram. The class A recording aperture is made in such form 
that the sloping edges modulate the light from the recording slit, but 
the noise-reduction slit is illuminated constantly, for signal ampli- 
tudes below that corresponding to 200 per cent of the track overload. 







FIG. 6. 

Method of transferring noise-reduction anticipation provided in the 
direct positive system to the master negative. 

The length of the image of the recording slit at the film is 76 mils, 
while the noise-reduction slit-image length is 100 mils. The shutter 
vane and the recording slit are the same length. When the modula- 
tion is near 100 per cent, the shutter takes the position shown by the 
dotted line in Fig. 5 A, giving maximum coverage of the noise-reduc- 
tion slit. The two ends of this slit remain uncovered, thus providing 
the black strips along each edge of the track. When there is no modu- 
lation, only the tips of the triangular shutter vane cover the noise-re- 
duction slit and provide the zero lines. The apex of each triangle is 
in line with the corresponding intersection of the recording light-beam 


and the modulation slit. It is apparent that the portion of the track 
beneath the uncovered center section of the noise-reduction slit re- 
ceives exposure from both slits. The portion of the track beneath 
the uncovered outer sections of the noise-reduction slit receives ex- 
posure from only this slit. The portion of the track beneath the 
covered sections of the noise-reduction slit receives either the full ex- 
posure from the recording slit or no exposure at all, depending on the 
position of the recording light-beam at any instant. 

Fig. 6 shows a method by which the noise-reduction anticipation 
provided in the direct positive system may be transferred to the 
master negative and therefore to the release print. A d-c amplifier 
with low-pass filter is utilized to make the current in the middle 
leg of the push-pull reproducing transformer control the motion of the 
shutter vanes in the re-recording optical system. This current is 
proportional to the motion of the shutter vanes on the direct positive 
optical system and is independent of the recorded sound-waves. The 
re-recording amplifier which drives the galvanometer, obtains its 
signal from the secondary of the push-pull transformer, which re- 
sponds to the recorded sound-waves but not to the shutter wave. 
Complete separation of the noise-reduction and the modulation 
waves is therefore accomplished without changing the time interval 
between them. If the recording film phonographs were provided with 
two reproducing optical systems spaced apart in the direction of the 
film motion, the final negative could be given the advantage of antici- 
pation by picking off the signal for the noise-reduction amplifier 
ahead of the regular scanning beam. The advantage of the method 
shown in Fig. 6 is that it can be applied to the present film phono- 
graphs or sound heads. 

The class B variable-area recording system is the simplest and 
most effective noise-reduction system yet devised. Direct positive 
recording opens new possibilities for the class B system by eliminating 
the principal source of difficulty, namely, the printer. Past ex- 
perience with class B recording under production conditions has 
shown considerable variation in quality, due to the effect of printer 
contact upon the resolution of the two narrow zero lines. The cor- 
rect rotational adjustment of the aperture to obtain a smooth cross- 
over between tracks was found to depend upon the condition of the 
particular printer. The direct positive class B recording tests made 
to date indicate that a smooth cross-over and therefore good quality 
may be obtained under varied laboratory and exposure conditions, 

486 G. L. DIMMICK AND A. C. BLANEY [j. s. M. p. E. 

At the present time it is not advisable to extend the direct positive 
method to include the standard type of variable-area track. Image 
spreading in variable-area sound records gives rise to a rectification 
component at high frequencies, which if not eliminated, results in 
sibilant distortion and a general loss in quality. High-frequency 
rectification may be brought under complete control in the printing 
process, because of the fact that the resultant distortion due to the 
negative and the print are of opposite signs and will completely cancel 
under certain conditions of negative and print density. High-fre- 
quency rectification also may be eliminated in the original recording 
by choosing the proper density, but for the films which are now avail- 
able, this cancellation density is too low to be usable. 2 The push-pull 
method is ideal for direct positive recording because the rectification 
component is eliminated in reproduction. This increases the proc- 
essing tolerances, and, with the absence of the printer variable, should 
considerably reduce the risks involved in obtaining high-quality 
original recordings. 

There are a number of ways in which a direct positive recording 
might be duplicated to provide a sound-track for the "dailies" and for 
editing purposes. For these purposes, it is assumed that the ultimate 
in quality and noise-reduction is not required. The original record 
could be re-recorded to another positive; it could be printed twice to 
obtain a positive; it could be printed once and reversed; or it could 
be printed once on the new auto-positive film and developed in the 
normal manner. The last method would be most desirable, provided 
this type of film could be made sensitive enough to be exposed in a 
commercial printer. For musical scoring, the direct positive system 
would lend itself to the present technic without any added complica- 
tions. Two recorders are nearly always used to cover a musical ses- 
sion because of the added protection. One of these could record a 
direct positive and the other a standard negative track. A print 
from the negative could be used for "dailies" and for editing, and the 
direct positive could be used for re-recording. 

A model of the direct positive optical system has been on test in 
Hollywood for a number of months. During this period all types 
of material have been recorded. Most of the recordings were made 
on the class A push-pull type of track; however, the class B push- 
pull proved to be practicable and, of course, offered greater noise re- 

Standard negative sound recording stocks were used. The ex- 


posed film was developed in either the sound negative or the print de- 
veloper, the exposure being adjusted to compensate for the difference 
in density speed of the solutions. There was no apparent quality 
difference with respect to the type of development. 

The exposure has proved to be much less critical than on the nega- 
tive-print process. Densities from 0.80 to 1.60 have been used, re- 
sulting in only very minor quality differences. This factor and the 
omission of the variations in printing no doubt account for the prac- 
tical performance of the class B track. 

In general, the chief benefit seems to be a lower noise level. A 
good print of a negative compares very favorably in quality with the 
direct record. However, since variations are always present in the 
printing operation, it is believed that a higher average of quality can 
be maintained from the direct record. This is, of course, advanta- 
geous from an operation standpoint. 


1 DIMMICK, G. L.: "The RCA Recording System and Its Adaptation to 
Various Types of Sound-Track," /. Soc. Mot. Pict. Eng. XXIX (Sept., 1937), p. 258. 

2 BAKER, J. O.: "Recording Tests on Some Recent High-Resolution Experi- 
mental Emulsions," /. Soc. Mot. Pict. Eng. XXX (Jan., 1938), p. 18. 


Summary. At the New York World's Fair 150 versions of a fifteen-minute 
story are carefully sorted to bring each to only four persons seated in comfortable 
chairs on a moving conveyor. 

A verbal description of a diorama along the edge of which the conveyor progresses, 
carefully synchronized with the motion of this conveyor, is given each group of persons 
and is repeated to each succeeding group with approximately a six-minute lag. In 
telling the fifteen-minute story, approximately 150 versions are being repeated simul- 
taneously, each version differing only in its starting time. 

In considering possible ways of meeting the elaborate requirements established for 
this sound system, various combinations of disk, film, and steel-tape reproducing 
apparatus were considered, a novel form of film reproducer being selected primarily 
on the basis of proved operating reliability over long periods of time. 

The Polyrhetor consists essentially of a rotating steel drum eight feet in diameter 
capable of carrying 24 continuous film loops past 168 optical scanners and associated 
amplifiers mounted on seven posts equally spaced about the drum. A multiple 
system of sectionalized trolleys conveys the sound through sliding contactors to small 
speakers in the cars, around which sufficient acoustical partitioning is provided to 
avoid program interference from car to car. 

The creation of a modern Babel might appear to be the purpose 
of the Polyrhetor, or 150-channel film reproducer, recently com- 
pleted for use at the World's Fair in New York. Actually, 150 
versions of a fifteen-minute story are carefully sorted to bring each 
to only four persons at a time seated in comfortable chairs on a moving 

The apparatus is a twenty- ton magnification of the "Call An- 
nouncer," the first model of which is satisfactorily operating in 
telephone plants after nine years of continuous service. The Poly- 
rhetor consists essentially of a rotating steel drum eight feet in 
diameter, machined to watch-like precision, capable of carrying 24 
continuous film loops past 168 optical scanners and associated ampli- 

* Presented at the 1939 Spring Meeting at Hollywood, Calif. ; received April 
17, 1939. 

** Electrical Research Products, Inc., New York, N. Y. 

t Western Electric Company, New York, N. Y. 


! fiers mounted on seven posts equally spaced about the drum. A 
i multiple system of sectionalized trolleys conveys the sound through 
I sliding contactors to small speakers in the cars, around which suf- 
i ficient acoustical partitioning is provided to avoid program inter- 
; ference from car to car. 

The need for this unique sound system came into being with 
i Norman Bel Geddes' concept of a gigantic diorama to portray the 
development of highway systems of the future and their influence on 
civic planning. Sponsored by the General Motors Company and 
known as Highways & Horizons, it provides the central theme in 
their New York World's Fair exhibit. Owing to the size of the 
panorama, over 35,000 square-feet, and the third of a mile of travel 
required to view it, a conveyor was needed to permit spectators to 
ride in comfort and view the scenes in proper sequence. To explain 
the significance of the panorama a verbal description and explanation 
was needed which must of necessity begin for each spectator at the 
same point of the ride and must at all times tell that part of the 
story appropriate to his position with respect to the diorama. The 
plans called for seats for approximately 600 spectators who are 
transported sidewise around the panorama. The conveyor took the 
form of 322 cars approximately 5 feet in length, coupled together in 
a continuous chain, each car carrying two chairs, except that every 
14th car carries, instead of chairs, an electric propulsion motor 
picking up power from a four-wire trolley system. The conveyor 
system is in continuous operation approximately 14 hours a day, 
access to and egress from it being obtained by means of moving plat- 
forms. Provisions were included for quickly stopping the conveyor 
in event of difficulty, and on being restarted, the sound must still 
be in correct relation to the spectators' positions. 

Many types of systems were reviewed in the preliminary study 
and three distinct methods were planned in some detail, before a 
final decision was made. These were, first, a series of self-contained 
disk systems, with automatic replaying attachments, housed under 
the chairs, on each alternate car, with a small amplifier serving four 
seats. Synchronizing pulses at the start of each panorama sequence 
would serve to maintain approximate synchronism. Second, a 
special steel tape reproducer was considered, having 150 pick-ups at 
intervals along a 1400-foot spirally wrapped tape. This would be 
mounted on the train and speech distributed through a flexible 
cable. The third plan was a film system at a central location. 


All considerations of maintenance, rapid service restoration, 
flexibility of program material, and ability to make announcements 
to the passengers, pointed to a central station installation with the 
minimum of apparatus on the cars. The most serious disadvantage 
to central station equipment appeared to be the fact that some 150 
communication channels must be provided from the central station 
to the individual groups of cars. Radio and guided carrier offered 
theoretical possibilities which, however, appeared impracticable. 
The logical method appeared to be some form of multiple trolley 
system with contactors mounted on each car. With the extremely 
limited space provided these parts would, of necessity, be so minute 
as to represent severe design difficulties and almost impossible 
maintenance requirements. 

In addition, conventional film systems involved moving contact 
with or flexure of the film, with consequent wear. However, a 
method is embodied in the Call Announcer 1 eliminating this dif- 
ficulty. In principle, this device consists of a series of small rotating 
drums, carrying sound-film past a conventional scanning system, the 
digits and party letters being repeated four times on each film. 
By a system of relays the desired sequence of digits is selected and 
repeated to the telephone operator in manually completing a dialed 
connection. Some of this apparatus, in continuous use for nine 
years in telephone plants, had repeated the digits over 100,000,000 
times with the same film and photoelectric cells without failure or 
replacement. However, the 15-minute playing time indicated an 
impossible drum diameter, which at 36 feet per minute would have 
been approximately 170 feet. 

Sectionalizing proved the key that unlocked the door to a solution 
which met all the requirements: merely sectionalize the length of 
the conveyor and the time of the sound. This adapted itself ad- 
mirably to the conveyor design, eliminated excessive numbers of 
trolleys, and required lengths of film which could conceivably be 
handled on the Call Announcer principle. From this sectionalizing 
method arose the first concept of the Polyrhetor. 

Spacing the traction motors at regular intervals throughout the 
conveyor divided it into a series of 23 sections, each containing 13 
cars with chairs and one with the motor. There are, therefore, only 
seven separate groups of spectators to receive sound in each section, 
six groups of four and one group of two, there being no sound required 
for the motor car. The cars being 5 feet long, established the length 

Nov., 1939] 



of each conveyor section as approximately seventy feet. Likewise, 
the time of sound in each section at a conveyor speed of 100 feet per 
minute is approximately forty seconds. 

On a continuous loop of film providing forty seconds of playing 
time, the story for the first seventy foot section of conveyor travel 
is recorded. A reproducer scanning this loop is associated with a 
length of trolley rail installed along the first seventy feet of panorama. 
The loud speaker of the first sound group in each successive con- 
veyor section is connected to this rail through a contactor drawn by 
the car. When properly synchronized, this contactor first touches 

FIG. 1. 

The Polyrhetor: General Motors exhibit at the New York World's 
Fair; Spectator Sound System. 

the trolley rail as the story is commencing, and as it finishes is car- 
ried from this trolley, entering a second trolley section, as the ap- 
propriate sound commences on a second loop of film. By spacing 
six more scanners (seven in all) equally around the loop, the start 
of the story reaches each scanner with a proportionate delay. These 
other six scanners are connected with trolley rails paralleling the 
first, and the following six sound groups in each conveyor section have 
contactors which progressively are carried into contact with the 
appropriate rail, just as the start of the sound reaches the associated 
scanner (Fig. 1). 


The Polyrhetor now took form. At a film speed of 3